Praxair, Inc. (NYSE:PX)
2013 Investor Day Conference
September 16, 2013 10:00 am ET
Kelcey E. Hoyt - Director of Investor Relations
Stephen F. Angel - Chairman, Chief Executive Officer and President
Eduardo F. Menezes - Executive Vice President and President of North American Industrial Gases
Scott W. Kaltrider - Former Vice President of Healthcare and President of Praxair Healthcare Services
John M. Panikar - President of Praxair Distribution Inc.
Matthew J. White - President of Praxair Canada
Antonio Cesar Miranda - President of Praxair Mexico and Central America
Scott E. Telesz - Executive Vice President
Todd A. Skare - President of Praxair Europe
Domingos Henrique Guimarães Bulus - Senior Vice President, President of Praxair South America and President of White Martins Gases Industriais Ltd.
Anne K. Roby - President of Praxair Asia and President of Global Electronics
Ray Roberge - Chief Technology Officer and Coprorate Vice President
Murray G. Covello - Vice President of Global Supply Systems
James S. Sawyer - Chief Financial Officer and Executive Vice President
Kevin W. McCarthy - BofA Merrill Lynch, Research Division
P. J. Juvekar - Citigroup Inc, Research Division
John Roberts - UBS Investment Bank, Research Division
Edward H. Yang - Oppenheimer & Co. Inc., Research Division
Robert A. Koort - Goldman Sachs Group Inc., Research Division
Donald Carson - Susquehanna Financial Group, LLLP, Research Division
John E. Roberts - The Buckingham Research Group Incorporated
Mark R. Gulley - BGC Partners, Inc., Research Division
Michael J. Harrison - First Analysis Securities Corporation, Research Division
Vincent Andrews - Morgan Stanley, Research Division
Kelcey E. Hoyt
Okay. Welcome. I'm Kelcey Hoyt, Director of Investor Relations for Praxair. We really appreciate everyone's attendance today at Praxair's 2013 Investor Day, everyone who's here live, as well
There are 3 exits. If you look toward the exit signs, they're kind of low. Those are the exits that we'll go through in the event of a fire drill. Okay?
As far as Q&A, if you'll take a look at your book, we have 2 groups of presentations today, a morning session and an afternoon, with Q&A time allocated at the end of those -- each one of those sessions. To the extent you are here and in person, you want to participate in the Q&A, I would just ask that you limit your questions to those presenters who are presenting either in the morning or save your questions for topics for the afternoon presenters in the afternoon. Okay?
For the lunch, we have a rotational-type lunch plan for today in an effort to give you the most exposure to the most number of Praxair businesses, functional folks, get as many of your questions answered as possible. So what we'll do is once we go through the presentations, we'll have the Q&A, take a quick break. And then here to my left, to you're right, is where we'll do the lunch. You'll notice that there are a series of tables that are numbered, and you can sit anywhere; it's not assigned seating. There are reserved seats, so those are for, typically, 1 or 2 Praxair folks. And what we'll do is rotate roughly about every 20 minutes during lunch. So it's a lot like speed dating. And to the extent that you're not enamored of 1 group of Praxair management, just wait, you've got 2 more coming through. Okay?
And then finally, cellphones. I would ask that you just take a look, make sure that they are either off or the ringer is turned off. I will warn you, I do have at least 1 member of management who, if the phone goes off, creates a disturbance, he might look at you, might invite you up on stage, ask you to perform, Scott Telesz.
And with that, I'm going to turn it over to our forward-looking statement. Please do read the slide. It applies to all presentations made today, as well as the Q&A session.
And with that, it's my pleasure to introduce our Chairman and CEO, Steve Angel.
Stephen F. Angel
Thank you, Kelcey. I want to welcome everyone here today. We're going to give you an in-depth review of the key aspects of our business. And I think we'll answer a lot of your questions along the way, but if not, we've allowed plenty of time to do so.
It's been 7 years since we've had a review of this depth, and I think you'll find it quite interesting. There are several things I want you to take from today. I want you to leave here with a deeper appreciation of the quality of our base business, the quality of our base business franchise. I want you to understand how we will grow our business in the future, what are the assumptions behind that and how we will lever that into earnings growth. I want you to meet the next level of Praxair management that has delivered our results to date and will deliver our results in the future. And lastly, I hope you'll learn something today about the industrial gas industry or Praxair that you didn't already know.
Now you're going to hear a lot today about discipline, execution, productivity, integrated supply. You will hear the words density mentioned a few times.
We're going to talk about the 3 Es or various aspects of the 3 Es, starting with emerging economies, and we think of that as Brazil, Russia, India, China, Mexico. Energy. We're going to talk about oil well services in Mexico, petrochemical and refining projects in the United States and Canada. I know there's a lot of interest in that. We're going to talk about gasification in China and how we think about gasification in China. And you will hear about the environment and how we will address environmental issues around the world, as well as meet the desire for higher standards of living in the emerging world.
We're going to talk about pricing and product management, something that you don't usually get a chance to see. And we're going to talk about how we manage our global energy spend. Now after people and benefits, our second-largest cost input is energy, predominantly electricity. And it's a $2 billion pie, and I think you'll find it interesting how we address that. And again, you're going to hear it all from the people who actually make it happen.
Now day 1 as CEO, which goes back some years ago, the board congratulates you, and then the first question they ask you is, "Where's your succession plan?" And obviously, I've had some time to think about that and work on that, and I continue to do so. But what I quickly realized is that the most important question, what has the most meaningful impact are who sits in those top 20 to 25 positions in the company. And to the extent that we can get that right, we'll be in great shape. So much of my time and that of the senior management team is dedicated to getting our top talent developed and prepared for those 25 or so roles and even greater roles down the road. Now at the end of the day, all I'm trying to do is get the right people in the right place at the right time for what's in their best interest from a development standpoint and for what's in the best interest of the company. And I believe that this group of people, many of whom are here today, is so strong and our operating system is so embedded in this company that you can eliminate the senior executive team, for a while, and the company would not miss a beat. And I don't want to see any nodding of heads from the front row.
Now there's several people I would like to introduce to you that are not on the formal agenda, but they do represent areas that might be of interest to you. And just kind of raise your hand whenever I call your name. Mark Murphy runs our Praxair Surface Technologies business. Mark's in the back of the room. And his biography, like everyone else from Praxair, is in your book for you to refer to. Amer Akhras is responsible for our Global Helium business. Amer? Dr. Riva Krut leads our sustainability efforts. And she just notified me, and I guess this is a public knowledge, if not, it is now. Okay, she says no, but it is now, that we've been recognized for the 11th consecutive year as part of the World Dow Jones Sustainability Index. And Keith Gordon is our Chief Operating Officer of NuCO2. He's in the back of the room there, and he's manning a NuCO2 display over in the other room that I hope you get a chance to stop by and take a look at. Sally Savoia heads up our Human Resources Department. Sally? All right. Sue Neumann heads up Communications and Media Relations. Sue? Jim Breedlove is our Chief General Counsel. Jim's in the front row. Liz Hirsch, who many of you know, is our Corporate Controller. Liz? Tim Heenan heads up Tax and Treasury. Tim? And Reynaldo Aloy works with Kelcey in Investor Relations. Reynaldo? All right, good.
Now I'm not someone that's big on history, at least not as it pertains to financial performance. But we are proud of our performance since our spin out in 1992. And what you see on this chart is our operating margins as a percent of sales compared to our 3 global competitors. And we have about a 600 basis point advantage over our nearest competitor in terms of operating profit as percent of sales. And if you refer to the graph in the upper right, you'll see that if anything, we've widened that advantage over time. And since the mid-to-late '90s, we've been able to expand our margin from 14% to 22%. Likewise, in return on capital, we have about a 200 basis points advantage today. And when you combine that level of profitability with the type of growth we've experienced over the years, it's not surprising that we've been able to grow our operating cash flow 12% per year since our spin out in 1992 to a record level of $2.8 billion in 2012.
And you'll note the takeaway at the bottom of the chart that our vision is to be the best-performing industrial gases company. Now that's nothing new. That's been our vision for a long time. And you will -- and when we think about best-performing, it's not just operating profit and return on capital. It's earnings per share growth, it's safety, it's productivity, it's working capital management, project execution, it's everything.
Now there's a document that we have in the other room that hopefully you'll get a chance to take a look at, but it talks about vision, mission, strategy, growth drivers, values. Vision, mission, strategy, growth drivers, which are the 3 Es, and values. And so why do we do this? Well, first of all, we like to get things on one page, which we've accomplished. But why do we do this? It's not necessarily for the Praxair people in this room. They understand all this. But it's important to get 27,000 people around the world aligned on what we're trying to do, how we will do it and what we stand for along the way.
So how do we create this competitive advantage? Now this is similar to a chart that we've shown before, except we've added some elements under execution and we've added a whole new column under people. And I think people make all the difference in the world. Now it really starts with select geographies. Before we go into a new geography, we spend a lot of time debating and deliberating that move before we pull the trigger. There's a lot of places around the world we're not in, as you know. We think we have the best footprint, and you'll see that profiled on the next page.
We very much like the core industrial gases business. We like the atmospherics business, we're not in chemicals or hygiene or sterilization or engineering and construction businesses. We love the atmospheric gases businesses. And we don't sell plants. We stick to the sale of gas model. That means we design, engineer, build, own, operate air separation plants under long-term contracts with high take-or-pay provisions for quality customers. That's what we do.
And we want to build out integrated supply systems, from on-site to merchant liquid, to packaged gas supply systems, to serve the full array of customers in that select geography, in that targeted geography. And we want to take advantage of the coproduct synergies that exist between those modes of supply. And Eduardo is going to come up right after me, and he's going to talk a little bit about integrated supply systems and the benefits.
And we want to build density to create economies of scale, optimize all the coproduct economics and leverage that installed base, leverage that infrastructure for future business at higher returns.
Now we have several technology advantages, the most significant of which in my mind is our portfolio of standardized product line plants. And this goes back to over a decade ago, where we had a lot of different plant offerings, we tended to customize for each project. And so we put a lot of work on consolidating around standardized product line offerings. And that has served us quite well over the years. Our hit rate for the projects that we go after and we don't go after everything, but the ones that we go after is about 60%, and our product line offerings is a -- an important reason why we've achieved that level of success. Now Ray Roberge, our Chief Technology Officer, will discuss our plans to expand our product line offering, as well as some of the excellent work his team is doing around application development to meet the needs of both the mature world and emerging world.
Now execution begins with operational discipline. We spend a lot of time on capital investment decisions in our company, a lot of time. And Ben Glazer will talk about that process, that capital investment decision process later on today. But just to give you an example, no one in our company can approve a project over $3 million without first reviewing it with me and receiving my approval.
Now safety, for those of you who grew up in operational background, you know that you simply cannot have good safety results and good safety practices unless your people are disciplined, unless you have operational discipline throughout your company. They go hand-in-hand.
Productivity. We simply cannot achieve the type of productivity that we achieve year in and year out without a strong commitment to continuous improvement, a strong ongoing commitment to continuous improvement. All the business leaders here in the room, and we're starting now to work in earnest on our 2014 plans, they know what the goal is. They know what the expectation is on productivity. It's the same as it was last year, the year before. It's the same, it's going to be in 2015 and 2016. It's 5% of their total cost stack. The goal does not change. And Sean Durbin will review how we make productivity sustainable in our company.
And I would say without a disciplined, rigorous process and senior management engagement, you'll never excel at pricing and contract management. There needs to be tension in the process, there needs to be a lot of discipline around the process, tension from a customer to the sales, sales to product management, product management to senior management. I think that's very important to achieving the kind of results we need to achieve. And Scott Kaltrider is going to talk about how we think about pricing and contract management, as well as energy management later this morning.
And of course, this probably goes without saying, but you cannot achieve top quintile performance in cost predictability and schedule attainment in your projects unless you have well-defined processes and the discipline to adhere to those processes. And you're going to hear from Murray Covello, who heads up our global supply systems organization, on that subject later today.
And the next one you might find odd that I put this under execution, and I don't mean for this to be a competitive advantage. I just -- I'm just trying to highlight the importance of it. But integrity and compliance. If you're a U.S.-based global multinational corporation with operations all over the world, which we are, you have to insist on the highest standards of business integrity from all 27,000 people that you have worldwide. And you have to make sure that you've got the compliant systems in place, robust compliant systems in place to ensure that compliance.
People. Everyone here today is here because they're a strong performer. They have a demonstrated track record of success. And not surprisingly, they're all strong operators. They excel at all those things that I listed under the execution column and many, many more. They're excellent people leaders, something that we insist on in Praxair. They're outstanding communicators. And they excel something very important to the senior executive team, they excel at driving alignment and accountability for results deep into their organizations. And I would say they're nonpolitical, certainly nonhierarchical. And if they had their druthers, I think the last thing they want to be doing is preparing PowerPoint presentations for some meeting. They would much rather be out working with the plants, working on safety, working on productivity, solving customer problems, et cetera, than preparing PowerPoint presentations. But they've done a very good job, and I think you'll see that later on today, and they're very proud of their businesses.
Now Scott Telesz, who's our newest member of the senior executive team, came with us in April of 2010. Right after lunch, he's going to talk a bit about his observations from working here for a few years, how we work the kind of people we have, et cetera. And I think you'll find his presentation very interesting.
So the takeaway from this chart is we're a high-performance culture. We certainly have room for improvement. We have operating systems, I'll say it again, deeply embedded in this company, and it takes years to build and it's not easy to copy.
So which should you expect from Praxair over the next 5 years, we'll call it? High single-digit sales growth is our expectation, about 8%. And you can see the breakdown of that in the upper right-hand part of the page.
We're expecting modest base volumes. As we look around the world, it's hard to expect that the next 10 years we're going to grow at the pace that, say, we grew up to the 10 years, up to -- prior to the recession of 2009. But we're assuming base volume growth, call it 3%. In that, I'm embedding contributions from our application development work. If the global economy does better, we'll certainly take full advantage of that. But this is our assumption.
Price in the 1% to 2% range. Historically, we've got about 2% price. And I think it's very fair to say that if volumes do a little better, the pricing environment is more favorable, and we'll certainly capitalize on that as well.
3% to 4% from projects, that's our call over this -- each year over the next 5 years. We think the backlog will trend down to around $2 billion, still healthy by historical standards and sufficient to support a 3% to 4% top line and bottom line growth from project contributions.
CapEx spend. We think going forward we'll be around $1.8 billion, plus or minus. That's about 13% of sales versus the, say, 20% of sales that we had in 2012, where we were at the top of the CapEx cycle.
I've included 1% for acquisitions. We have a strategy to consolidate the packaged gas industry in the United States. So we're going to be spending some money towards that, call it $100 million to $200 million per year. Nothing on the order of NuCO2. I think it could be a decade before we do anything of that size. But this is our plan.
Operating profit will grow faster than sales, and this is from pricing and productivity. The way we think about it is if you have $100 of inflation, you need at least $200 between pricing and productivity to maintain operating leverage.
A few years ago, when we were at -- I don't remember exactly the operating margin percentage, I'll call it 20% to 21%, somebody asked, "You're industry-leading today, can you really continue to grow that?" And my answer was yes. And this is how we grow it. I said I think we'll grow at 30 to 50 basis points per year going forward, and I still believe that. And this is how we'll do it.
Our plans going forward are to reduce the share count about 2% per year. Now this year is a bit of an anomaly. We said it'd be closer to 1% because of the NuCO2 acquisition. But going forward, we believe 2% per year share count reduction is the right place to be. And the combination of all that will get us into double-digit earnings per share territory.
Return on capital is about 13% today. We think we can move that to 14% and make it back to the 15% that we were in the mid-to-late first part of the decade. How will we do that? Well, we're going to start to get the payback, and we are starting to get the payback as I speak, from the big CapEx spend from 2012 in the first half of this year, and those projects will ramp up and contribute more as we go forward. Obviously, the NuCO2 acquisition at the beginning of the year was dilutive to our return on capital. But as that business continues to do well, and it is doing well, very well, we expect that dilutive effect to lessen over time. Obviously, CapEx spend moderating has some contribution, not a lot, but some contribution to a higher return on capital. But we need base economic growth to get to 15% for sure. We have -- if you look around the world, we have underutilized capacity. Europe is a prime example of that. But we have it really all over the world. And with some modest growth on a continual basis, without any additional capital to meet that volume growth, this will enable return on capital to continue to march forward. And again, 15% is our goal, and that's what we're going to get back to.
So what could go wrong with this forecast? I think I just alluded to it. No base volume growth, no global industrial production expansion. But it is built into our assumption. Now we'll still get some growth from applications, from hydrogen, from gasification, from decaps, projects that are not tied to industrial capacity expansion. But without some volume growth, it's extremely difficult to deliver double-digit earnings per share growth. And there's one other thing that could go wrong, my assumption is -- assumes stable currency. Clearly, we've had some problems with currency last year. This year, we still got some headwind. But if I face too much currency headwind, if we face too much currency headwind, it'd be very difficult to get to double-digit earnings per share growth. But the things that we can control, productivity, cost, pricing to a large extent, project execution, working capital, all those things that we can control, we will.
Now what will our sales profile look like at the end of 5 years? And this is our best estimate. We think North America and South America will still be about the same level of representation in the sales mix as they are today. North America, about 50%; South America in the high teens. I think the U.S. and Mexico will grow at pretty attractive levels, something on the order of what I'm expecting globally, around 8%. Projects will certainly play a role in that. I think Canada will lag that somewhat, and I'll let Matt talk a little bit about that later. I think Brazil, short-term, will probably in the mid-single digits; longer-term, still has a lot of potential. Europe will drop from about 13% of our sales today to around 11%. We're just not anticipating much growth out of Europe over the next 5 years. And Asia will make the biggest move from about 12% of sales today to around 16%, 17% over this 5-year time frame, really driven by the growth of China, which is -- a lot to talk about China. It's 4% to 5% of our portfolio today, and we think that will grow probably to something on the order of 7% to 8% over this time frame. And you'll hear more from Anne on that subject.
End markets, food and beverage is up -- is a little stronger, was 6%, now it's about 8% with the NuCO2 acquisition. Energy makes the biggest move, from about 11% to 14%, and that is really on the strength of hydrogen coming out of the backlog. And our hydrogen sales will double over this time frame from our 2012 levels, and Dan Yankowski will talk a little bit about that.
Supply mode, roughly 25% or so on-site business today, we'll push closer to 30%, again, largely on the strength of the backlog.
And how is this different from our competitors? I'm sure some of you are sitting there thinking that. Well, we have the largest presence in North America and South America today. That's a differentiator. We have strong franchises, top to bottom every country. And we're very focused on traditional markets, manufacturing, metals, energy, chemicals, traditional markets. We have a very small presence in health care, and that presence is smaller today than it used to be. If you remember, we divested the U.S. Homecare business. This differs from our competitors, from our European competitors who are investing heavily in the health care business globally.
And lastly, we have the smallest presence of all the global competitors in electronics. So if you like the Americas, if you like the traditional markets segment, the manufacturing, metals, chemicals, energy, we're your company.
So to kind of just tie it all together, we expect solid growth, solid profit generation over the next 5 years.
And with that, I'd like to turn it over to Eduardo Menezes, who will -- who's our Executive Vice President of North America. And Eduardo is going to talk about how we think about organization, how we do organize and again, a little tutorial on integrated supply. So Eduardo?
Eduardo F. Menezes
My presentation today has 2 main objectives: one, to explain how we organize Praxair operations in a global level; and second, to take a deeper dive into our U.S. organization and explain the advantages that we see in our integrated business model for the U.S. vis-à-vis our competitors.
So starting with organization. There is basically 4 major dimensions you can use to organize an industrial gas business. The first one is product. We have a lot of products in our portfolio, if you think about specialty gases and mixtures. But the reality is, 6 major projects -- products make more than 80% of the sales of this industry. So the 3 main, atmospheric Gases: Oxygen, nitrogen and argon, plus hydrogen, CO2 and helium. It's a viable way to organize your business for a product that has a Global Supply Chain such as helium. But really for the other products, the differences that you have from region to region and especially, the differences that you have in supply mode that I'm going to explain next, are so big that it's not a good alternative in a global level and a regional level. The second option you have to organize your business is by supply mode. And the one interesting characteristic of our industry is that we supply our end customers, regardless of their size. So and in a scale that I challenge you to think about the different -- not a product in the same scale. So just to use oxygen, as an example, we may have customers that take -- anywhere for like 5 pounds a day of product, like a patient that needs oxygen for respiratory care or a welding shop, all the way to 5 million pounds a day of product for a steel mill or a petrochemical company and sometimes even more than that. So the scale is like 1 through 1 million. And it really causes you to have different ways to supply these products.
So if we have, staying with oxygen, staying with the #5, if we have a customer that requires more than 50 tons a day of product, for example, we're probably going to install a plant close to the customer with prolonged side supply mode, and we're going to connect the customer with the pipeline and supply gaseous product to this customer. And the name of the game here is really how well you design, build and operate the plants. So it's an engineering and operations business, as Steve was describing. If the customer needs less than 50 tons a day, all the way down to probably 50 pounds a day, we'll probably supply the customer using liquid product. So liquid product comes from the same plant that we used to supply the pipeline customers, and we -- where we liquefy the product, we store at very large tanks, and we bring the product to the customer using the trailers that you can see on the pictures on the back of this room. And at the customer, we have liquid tanks where we store the product and again, we vaporize it back to the gaseous form because that's normally how the product -- the customer use the product. It's a business that -- it's all about logistics, and all about how well you can manage at a very tight supply chain because at the end of the day you have very few days of field employed in the entire supply chain for liquid products. If the customer needs less than 50 pounds a day, then the only solution you have is a high pressure cylinder. Now high pressure cylinder is not a very convenient way to transport a gas, a cylinder can weigh more than 250 pounds of steel, and you cannot carry more than 30 pounds of product inside this cylinder. And on top of that, the cylinder's worth more than the product that you have inside of the cylinder, most of the time.
So the name of the game here is how well you manage these assets, a company like Praxair has more than 10 million cylinders. So how well you can move these cylinders to the customer, bring them back, dance with you and so forth. And it's a very labor-intensive business. So very different business models, so you can think about organizing our business around supply mode. But the reality is, especially for packaged gases and for merchant, that of very regional business, you may start to get in trouble if you try to manage for the business. We, like that, we do not believe that we should have someone in our headquarters in Connecticut, trying to manage Packaged Gases businesses in Buenos Aires, Beijing or Berlin. And the person would not really be effective, trying to do that from our headquarters. But it's a valid way, and I'm going to talk about how we do that in a different level. The third dimension that we have is by end user. We have some industries that really have very specific periods, requirements or quality control requirements or they need special equipment in order to dispense the gas. And you can think about organizing a business around that, and you see a lot of that in the industry with electronics, with health care. At the end of the day, it's a viable alternative. But really, suboptimize the assets, as you need to dedicate assets and people for a certain industry, and you cannot share with other markets, it really gives a lot of purchase power to your customer. So -- but it's a valid option. And the fourth option is the geographical option, is selling the simplest one. It is required at a certain level, you always need to have a national organization, you need to file taxes, you need to follow regulations. So if you organize yourself around that, you save some money. And it really helps you, take in consideration all the competitive intensity that you have in each market, and helps you enforce P&L accountability. But on the other hand, requires very strong general management skills to run a business with all the products that I described, all the supply modes I talked about, and all the end-users and have one person managing everything, it's not an easy task.
At the end of the day, you need to pick a model. You need to maintain discipline and you need to manage the exceptions. So how we organize our business at Praxair?
We have no shame about that. We prioritize P&L accountability. We try to keep it simple and we try to run as a regional business model. So if you look at the way we report our results to you every quarter, it's exactly the way we organize our business. We have 4 main regions: North America, South America, Asia and Europe. And underneath these regions, we have countries. And the country managers in Praxair, they are the final accountables for P&L in our company. So they run sales, they run operations, they run distribution, they are in charge of optimizing asset utilization, and they're in charge of maximizing the ROC, which will, at the end, compound to the ROC that Steve's talked about before.
Now underneath that level, in the country level, we try to organize that it makes sense as the conditions for the local market dictates. And the next slide I'm going to talk to you about, the U.S. organization, which is by far our largest and most complex organization. And with that, I hope you can have an understanding, good understanding of how we organize the other countries around the world.
What are the advantage that we see in our model? We believe that allow us to focus on the results, really be fast responding to market changes, to our customer needs and we -- especially, we avoid the conflicts and the distractions that you have in a matrix organization, really try to avoid any kind of matrix feeling in our organization. And also give us a little more flexibility. One presentation you're not going to be see in your books today is about some big ERP implementation at Praxair, most likely a lot of people here do not know what the ERP system or systems we use at Praxair. And the reason is simple. We -- although we have a global platform, and most countries follow that, we are comfortable running separate P&Ls in separate countries in different systems, and also to bring acquisitions using different systems, which allow us to basically be much more focused on the business and not focus on IT transformation. So we like this business. We like this format, and next, I'm going to explain what we do in U.S. and how we run the business in our country here.
So this is a little complex slide when I put that together, people ask me how we're going to try to explain that. But I'll go little by little here, so hopefully we'll get there. First thing you'll notice on the top, on the blue bars, you'll see that by supply mode, our division in the United States is very similar to what Steve showed in a global level, and very similar to a 1/3, 1/3, 1/3 between on-site, merchant and packaged. Underneath that, you'll see several P&Ls that, individual P&L's that create the national P&L for the U.S. And the green ones are regional P&Ls and the gray ones are national P&Ls. So how we manage our business in U.S.? The first thing we do, we break apart the Packaged Gas business. And we do that for the simple reason that the Packaged Gas business in the U.S. is unique. We have more than 700 distributors, independent distributors that we need to compete with. We have a market that is used to buy what call hardgoods or welding supplies from the gas suppliers. So you need to have a network of warehouses and stores, so it's a very different business that we need to run across a structure that is competitive with these 700 distributors that we need to face in the marketplace. So in order to do that, we create a separate subsidiary that we call Praxair Distribution Inc. or PDI, and that John Panikar, that runs this business will come later this morning to talk about how we run our Packaged Gas business. But without stealing his thunder, 80% of the business we run in a geographical mode. So we believe that this is a local business, so we break this apart in 4 regions and 12 divisions. And every division has its own General Manager that is in charge of the P&L and run the business from sales to the bottom line.
In addition to the regional divisions, we have 2 segments that we think are significantly different from the day-to-day business on industrial packaged gas and hard goods, and they are spec gases in the institutional healthcare that we run with the national P&L always -- all reporting to John and closing the P&L for Praxair Distribution of PDI.
On the other side of this slide, you'll see our On-site and Merchant business. We run that -- internally we call that United States industrial gases or USIG. And that you see what we do in this business, we basically, we take 60%, 70% of the business that is the basic Atmospheric business, both in On-site and in Merchant, plus the liquid CO2 business and the liquid hydrogen business, through which, very similar to the distribution systems that we have for oxygen, nitrogen and argon, and then we combine them in 3 regions. So we have 3 region P&Ls that have run for 3 VPs in U.S. And really, we try to avoid any discussion on allocation of costs as we've been an air separation plant, so the cost belongs to the on-site side or the merchant side. It's all run within the same region, within the same PDI -- or the same P&L. We also have 4 national P&Ls that you can see here. We have the on-site Hydrogen business that, it's very different in terms of operations and business development. So we have a dedicated group running that business. So we have 1 for oil and gas services, we have 1 for NuCO2 that now became part of this organization. And we have the Helium business that, it's run out of the U.S. but it's really a global business from the U.S. We control all the -- all the logistics, all the purchase of product that we buy from third parties. And the -- and really, the supply to main customers in a global level, although our subsidiaries around the world, the countries, they still receive liquid from the U.S. and they pumped and they supply local packaged gas customers and local small liquid customers regionally.
So at the end of the day, we've run all these P&Ls, but we are also very careful on trying to maintain consistency on how we run the business, try to avoid duplication of costs and try to replicate good practices. And because of that, we have this centralized functions that you see here in the bottom. As Steve said, we're going to have 2 presentations following this one. Talking about the centralized functions. The first one will be about Product Management and Energy Management, which basically is a group that we have trying to make sure that we are consistent across the country on how we price products, how we expand capacity and so forth. And the other one within our operations group. We'll talk about productivity and how we move productivity across the regions without duplicating costs in the organizations.
So this structure that I just described to you is what allow us to run a competitive packaged gas business. We've been a major industrial gas company, and this is a unique situation. We are the only fully integrated player in the United States, as you can see here in these pictures. And that when we go to the market, when we face our competitors, we believe we have some advantages. So versus the automated industrial gas companies that do not have packaged gases in U.S., we believe we have the advantage of maximizing the core product economics. So when we view the plant, we consider not only the liquid that can come out of this plant, but also the Packaged Gas business. We have access to a much larger sales force that comes with the business, so we have more than 700 people in our packaged gas business in the sales side, and we have the ability to grow with the customer. A lot of times, the customers, they start with packaged gases and they move to do doers, they move to micro bulk, and finally move to bulk. And if you are in the packaged gas business, you have the ability to grow with the customer, and that's an advantage that we have vis-à-vis our competitors in the U.S.
On the other side, versus the packaged gas is oriented competitors, we believe we also have advantages we can -- we have the -- we source our own products, which is very important, especially for products like helium and argon. So the customer, they are always sure that we'll have the product because we control the entire supply chain. And we also have a greater percentage of their wallet share and we are able to supply these customers on a national and international basis, which give us an advantage when we talk to customers about packaged gases. So we believe that this model is what allow us to be, not only the largest, but also the best-performing industrial gas company in the U.S. So this is all I have. I'll be happy to answer any questions you have during the Q&A or during lunchtime. And after that, we'll have Scott Kaltrider, which is the VP for business management in USIG, in addition to product management, and Energy Management, Scott is in charge of national P&Ls for both helium and oil well sources. Scott?
Scott W. Kaltrider
Thanks, Eduardo. Good morning. It's great to be here. It's good to have all of you here to talk about Praxair a little bit this morning. I'm going to speak specifically this morning about our product management and energy management functions, and specifically around product management, I'm going to talk about what our approach is to contracting strategy in contract management, and how that plays into our overall pricing approach and our pricing program. And I'm going to have some comments around our Energy Management, because we feel like we have to be world-class at that. We think we are world-class at that. As Steve mentioned, it is our second largest spend as a percentage of our overall cost stack.
So if you look at this chart, this chart represents the key product lines that U.S. Industrial Gases manufactures and distributes. And as I put this chart together, as we develop this presentation, it struck me as a slate of products that are pretty diverse and complex. And that may seem strange, because when you think about cryogenics, you say, well, there's a lot of similarities, and while maybe the end products are similar in cryogenic nature, how you get there is not. And so there are some very substantial differences across these product lines. And specifically, I talk about manufacturing and production and distribution characteristics, vary widely across these product lines.
The degree to which you can control your sources and your feedstocks, there's a large variation across these product lines, has a big impact on your sourcing strategy. And thirdly, and maybe most importantly, the markets associated with these products and the scope of these markets span local to regional markets, in the case of your typical oxygen, nitrogen atmospherics, down through national markets and how we look and view these products from a national basis, and that includes argon, hydrogen, CO2, and then all the way into global markets, which Eduardo talked about, with helium and rare gas. And its because of these differences and these vast differences in the makeup, in these characteristics of these product lines, that Praxair has really taken the approach to dedicate product managers to each of these product lines. And so in Praxair, we have a national product manager that looks after the profitability of oxygen and nitrogen. We have a product manager specific to argon, specific to hydrogen, specific to carbon dioxide, and then we have a product manager that looks after our helium and rare gas business. And these product managers are centrally located in our headquarters in Danbury, and we found that, that's the most effective way to really, truly manage the profitability of these product lines. Because you're able to get discipline and consistent processes applied throughout the product line, so your approach -- in other words, your approach and your processes of managing each product line are consistent, but you're also able to realize synergies, to the extent they do exist across the product lines.
So that's how we're organized, and we feel that, that's worked well for us. Let's talk a little bit now about how the product management function shapes our contracting strategy. As Steve mentioned in his opening remarks, you really can't overstate the importance of a solid, outstanding contract management strategy. It really is the underpinning of your whole price program. If you're not outstanding and excellent at negotiating agreements that provide the terms and conditions, that allow you to take advantage of market-based changes as well as recover your cost, you're not going to be successful in the pricing arena.
So for many of you that follow the industry, these are very traditional supply modes: On-site, merchant, packaged gases. And you can see the continuum that starts at the top, migrates down, high capital intensity, high-volume, medium capital intensity, kind of medium contract term, and then down into the package supply mode, relatively short contract term and a lot of transactions and a lot of flexibility required. And so one, of the things I'd like for you to take away today is, no matter what product mode, supply mode or customer we're supplying, one of our main goals in our contract managing strategy, is to obtain either cost pass-through or very, very high degree of cost recovery. And to the extent you move down this continuum, and you look at the merchant and packaged business, we like to include specific terms and conditions that allow us to take advantage of changing market conditions, so that we can get market base price adjustment. Now I'll talk a little bit more in specifics about each mode of supply, and what we're trying to achieve, because there are some obvious differences in an on-site project, and specific terms and conditions you want to achieve in an on-site contract, as opposed to the other supply modes. So as many of you are aware, an on-site contract is typically associated with a very high capital investment, usually a large plant, either on a customer's property or adjacent to the customer's property. Manufacture the product, delivered through a pipeline. So these are long-term agreements. You don't have a lot of opportunity during that contract term to fix it, so you've got to get it right upfront.
And so Praxair, as many of you know, are very selective on the projects we take. And that happens to -- is borne out in our return on capital, and the returns that we get. And so when we do take a project on, we want to make darn sure that, that return, the expected return, is delivered throughout the project life. And so there's 2 key provisions that we look for in every on-site agreement. And Steve mentioned this as well. We're insistent on a take-or-pay provision. And you know it's funny because, when things are going really well in the economy in gangbusters, and -- I've seen firsthand where we've had competitors take on, and we've walked away from projects where we couldn't get a take-or-pay provision. And when you look at the conditions to what happened out there after the crash of '08 and into '09, some of those projects don't look so good today. So we are very sticklers, very much sticklers for the take-or-pay provision. And it ensures a base return for us, a base return on our project. And then the second thing is, like I said, we want to have formulas and escalation built in that pass cost through to our customer, through construction, through start up and through the total economic life of that project, so that our return is not degraded over the term of that project.
When you move to the merchant contract, it's a medium term contract, 3 to 7 years, typically. They are requirements contracts, which means the customers are required to buy product from us for the duration of that term. And this is probably -- or this is a very wide range of customers. You can have customers in this category that take 50,000 cubic feet a month of product, and you can have customers in this category that take 50 million, or 100 million feet of product. So you can imagine the range of sophistication of the customer that you're dealing with, varies widely. And what you're able to negotiate varies widely. But again, what are we looking for? High degree of cost recovery, and terms and conditions that allow us to take advantage of market-based price adjustment, when possible.
So what we really try to do in the Merchant business is we employ a portfolio approach to our contract management, and we have contracts that are open, we have contracts that are formula. But in all cases, in almost all cases, it's never an absolute, in almost all cases, we incorporate terms and conditions which anticipate changing market conditions. And finally, you get to the packaged gas business, and the packaged gas customer, they're short term, relatively short term contracts and there's one thing again, we're very insistent on, in the Package business. We get paid for rent on our cylinders that are out at the site, and to the extent we have equipment like mixing panels, gas mixers, manifolds, whatever that is, we're going to get a facility fee for that equipment. And then we want to maximize the flexibility of these agreements, because when you're dealing with very short-term agreements, highly volatile market conditions, lots of transactions, you need maximum flexibility. So in summary, on this slide, #1, what we're shooting for, regardless of the supply mode, is cost pass-through, or extremely high degree of cost recovery and then tailoring our Ts and Cs to the specific supply mode.
Talk about pricing. This is very hard work. Pricing does not come easy and does not happen by itself. And you need total commitment in your organization, from the very top, out into the field organization. And Steve used the word tension. And that's not a bad word, because you have to be committed as an organization. You have to be disciplined, and you have to be prepared. I can tell you, I've been out there, I'd been across the desk from the customer many times. And when I say prepared, I mean, you really ought to know as much about that customer's business as he knows, if you want to be successful. And so as we look at the merchant pricing arena, there are some very favorable fundamentals that lend themselves to being able to do some pretty good things in the Merchant business. And the first of these is, and it's fortunate is, in general, our products represent a low percentage of our customer's cost stack. On the other hand, the vast majority of our customers can operate without our product, which means you rely -- your reliability better be superior. And in the end, our products really don't travel very well. They don't have a good shelf life. So if you're going to be in this business, you're going to have to invest a lot of capital to compete, and that's an inherent barrier to entry. So how does Praxair view pricing and what's our approach?
With us, it really begins with reliability. Reliability is king. And I'm talking about system reliability. Your entire system has to be reliable. And so Praxair is very, very intent on reinvesting in its infrastructure, and reinvesting in technology to ensure our reliability is second to none in the industry. And I know Sean is going to come up next and he's going to talk about reliability. What else? You have to bring value to your customer over and above the intrinsic value of the product you're supplying. Fortunately for us, at least at Praxair, we have lots of ways to do that. We have applications technologies that can improve yield at a customer site, that can improve productivity, production, can improve his quality, you can take on other -- if you improve quality, it opens up new markets. We have productivity programs that we apply internally in our company to drive productivity. We can take those to the customer who doesn't have a productivity program and eliminate waste. There's lots of ways that we can bring value to our customers. And if you're successful being the most reliable in bringing additional value to your customer, you create a very positive environment for your contract discussion. If you have a positive environment for your contract discussion, you're going to get the majority of terms and conditions that you need to get in order to be successful. In a high degree of cost recovery, and if you get a high degree of cost recovery, when you are able to implement a market-based price action, you're going to realize the majority of that price. And that's the result that you see on this graph to the right. What this graph to the right really shows is, over the last 5 years, and we all know we've been through a lot of economic cycle in the last 5 years, Praxair's been able to grind out, in the U.S. and Canada, pretty much a 2% on a CAGR basis of price. And that's not easy. And we feel pretty good about that. Now we're not perfect. We'd like to be able to do better. But that's something that we feel like we can count on. And that's embedded in our people and in our culture.
Okay. Last thing I want to talk about, energy and energy management, and again, as Steve mentioned, $2 billion of energy spend globally, between power and natural gas. I'll call your attention to this graph up in the upper left quadrant, the northwest quadrant, if you will. Inherently, the ASU is a very attractive load for a utility. High megawatts, high load factor, equals high cost recovery for the utilities. So they really like that. At Praxair, we look at some specialized design features of our plants. We modularize our liquefaction. We specify equipment that has wide-range ability performance without efficiency degradation. We look at the storage capacity, make sure we have the right storage capacity, which allows you to take advantage of peak and off-peak lower power rates and to make liquid during those times. And if you're able to do that, you can shape your load, and if you can shape your load, you can take power at those periods of time during the day that are cheapest. You take that to a further extent and you go down here to the southwest quadrant, and you can actually drop load. Okay? So on a very short notification from the utility, if you've designed your facility right and you're working right, you can actually drop load. And that allows you to procure interruptible supply power, which saves you, frankly, millions. And you layer that interruptible in, in place of your firm power. Okay? And the result of all this is a portfolio approach that looks like this pie chart you see. And the conclusion really is that you can enhance your profitability of your merchant business dramatically without sacrificing the reliability of your operations.
So that's what I have this morning. I look forward to talking to you at the break. I think Sean Durbin is up next. Sean is our Vice President in USIG, in operations. And he's going to talk about reliability and productivity. Thanks a lot.
Good morning. I'm going to talk about productivity and really how it's connected to Praxair being a reliable supplier to our customers. Now we started down the productivity path and it's current generation over 10 years ago. And through the years it's evolved and developed into really being a key part of our business model today. But more importantly, it's a key part of the culture of the company. Over 27,000 employees understand its importance, understand that they're going to be involved, and they do that day in, day out. Now we have an internal mantra in the company around productivity and that we're not perfect so there's opportunity.
And people keep that in the back of their mind, and it just helps them understand that we need to continue to improve every day and make sure we're delivering results.
I'm going to start by just giving an overview of how productivity is weaved into our business. Essentially, our productivity model is based around 3 functions. You have the front lines, the sales professionals, the plant technicians, the accounting department, they really make the business run on a day-to-day business. Now, Steve talked about density being a strategic focus for the company. In terms of the U.S. what that means is really the several hundreds of plants we operate and the several hundreds of thousands of deliveries we make every year. And when you think that level of transaction, that level of interface, you can think about the amount of data that you generated. And so we have support groups that their job is really to analyze that data, understand what's happening, identify anomalies where we may have an underperforming piece of equipment, maybe an underperforming region, understand what's happening, address it and move it up to the average. As important is understanding what the best performers are doing, where we have a plant that may have a higher efficiency than its peer group. How are they doing it? Identify it and replicate that among its peers as quickly as we can. And what does that does is basically allow you to move up the average.
Now the key is you identify the opportunity just as important is to actually get it done. And so what we have as really the third function, is the dedicated productivity teams. So these teams really take the ideas, and those ideas can come from anywhere, they may come from the frontline, they may come from the support groups, they may come from a different country. They take that idea, develop it into a project, get that project executed in a timely manner, and then most important, make sure that we get the benefits realized and those benefits are sustainable. In a lot of ways they're like a sales group because they have a backlog of opportunities, they have specific targets that they need to achieve and we measure them on the amount of value that they create. So again, it's the robust analysis and execution that really drives our continuous productivity.
In terms of -- I'm going to give a couple examples. I'm going to talk a little bit about distribution, a little bit about production, because when we look at those areas, they're part of what we call our value streams. And really, our value streams are: distribution, production, reliability and business processes. Within distribution, it's a very labor-intensive process. You need to plan the tour and you need to get the product to the customers. And getting that done day in, day out, when on average, in the U.S., we're making the delivery every 90 seconds. Again, you can think about the number of interactions that come into play. Demand, equipment reliability, even the weather come into play. And so the key to the distribution costs is really driving down your cost per mile and driving up your volume per trip. And so in terms of our distribution costs, fuel and labor are the largest components. And we spent a lot of time making sure we're improving those areas. And a lot of the new technology that's coming out is really helping us do that. For example, we've just finished installing our next-generation onboard computers. These computers are mounted in every truck. They give us detailed information about driver behavior, things like breaking, acceleration, idle time. Those factors have direct impact on fuel economy. And you can imagine, when you can improve driving behavior, again, move the underperformers up but move the entire group up, you're making improvements in fuel economy and you can see what sort of impact that would have on our distribution costs.
Similar, it's about maximizing the resources. So for us, we want the drivers getting our product from our plant to the customer. And so any time they spend at the fill plant, waiting for the truck to be filled, or at the customer, waiting for that product to be unloaded, is time they're not behind the wheel, and so advances in control automations, pump technology, really are minimizing the amount of site time the drivers have to spend, which again puts them behind the wheel and allows us to be more flexible and nimble for our customers.
On the volume side, it's important because we're dealing with products, in many cases, that are extremely cold, right? And so it's important you get as much product on that truck as you can, and you get all that product off the truck before it rides back at the plant. Otherwise, you're going to do more trips than you need to do. And so each of our tours we actually grade against a rating scale. And it follows the same approach. Where are we doing it well? Where are we doing it not so well. We identify what the trends are behind that and address them quickly. Now on top of that, there is an optimal delivery window that we're trying to target. And we have an internally developed logistics system that we're actually in the process of upgrading right now. And what that lets us do is really position the trucks and our resources so we're delivering the product right when we need to. You obviously don't want it deliver early, or that's an inefficient delivery and that's driving up your cost. And you absolutely don't want them to deliver late because that affects the reliability to the customer. We're now taking these concepts and we're actually migrating them outside of our bulk business, into areas like microbulk, even NuCO2. For example, NuCO2 has a scheduled delivery system, where if you're a particular cluster, you may get your delivery every Tuesday or every Wednesday. We're migrating that to a inventory-managed system. And again, what that allows us to do is really match the supply to the demand, which reduces our costs. But, as important, it makes sure that product is always there when they need it. So it improves customer reliability.
On the production side, we have several hundreds plants we operate in the U.S. And there's a multiple of that around the world. And when you look at the cost of production, energy really is the single largest component. And most of that energy is consumed by compression and rotating equipment. So over the years, while there are commercially available equipment suppliers, and we use them quite regularly in our plants, we've developed the internal capability to basically retrofit those -- that equipment and, in some cases, build entire pieces of equipment ourselves, which really maximizes the efficiency. And that tradeoff is really dependent by application, by vendor. We don't need to build things just to build them. So if we can work with an OEM, that'll be our first approach. They'll take our technology, apply it to their equipment for our use. In other cases, the efficiency improvement and really the drive in lowering the total cost of ownership makes it better if we build the entire piece of equipment.
So these are some examples. And if you look at it, the key point here is these projects have been done in the last -- this year and last year, is the ability to replicate these projects. And that's really a benefit of having a single focus on our business, we're able to take these ideas, replicate it not only in the U.S. but actually around the world. And the takeaway on this slide, I think, is important because, to use the baseball analogy, it really is a game of singles. I'm standing up here as a representative of operations for the U.S., but my counterparts around the world could be on this stage delivering you the same presentation because they're doing the same thing every day. We have thousands of people doing thousands of projects, delivering the productivity. And it's really that granular approach that we believe is the differentiator. We don't have a special name for our productivity program. We don't even call it a program, it's just what we do. It's part of the DNA of the company.
Finally, some of you have seen this slide previously. And I talked about some of these areas. And you'll hear about the other areas throughout the day in other presentations. As Steve mentioned, there's 5% cost stack reduction target. It's there every year. Every business has it. Every region has it. We know where we stand against that target. We know where we stand against each other. And really, that just makes productivity not be an annual thing you do. Say, okay, I need to get this done at the beginning of the year. You're just working on it all the time. And again, when you go back to the granular approach we take, it's not something where you have a big project and it's a one and done, it's all 27,000 employees doing it every day all the time. We get asked all the time can -- how can Praxair sustain this level of productivity benefits? And really, the answer is we can for really 3 reasons: first, we have a proven methodology. We know how to get it done. We don't talk about what we like to do. When we have an idea, we define it and we get it done. If there's not capital expenditures involved, our target to get that done in 90 days and move on to the next thing. We have a results-driven culture. When we put our minds to it and we want to get it done, we get it done, and that's all of our employees. That's what we preach within the company, that's how we develop our people and that's really what our results have shown. But most importantly, there's just a lot of opportunities. Again, when you're looking at every aspect of the business and you have all the employees looking at that and may feel rewarded for coming up with those ideas, we'll have a technician come with an idea to do something different. We'll save money at that plant, we'll then replicate that within the U.S. We'll then replicate that outside the U.S. and other countries. Their performance is recognized. We'll have members of the senior executive team meet, get to know them. What that does is suddenly you go from knowing people's name to being on a first name basis all the way up and down the organization. And that helps keep the motivation going. So again, our view is productivity is sustainable.
And with that, I'd like to turn it over to John Panikar, President of Praxair Distribution, who is going to talk about our packaged gas business. Thanks.
John M. Panikar
Thank you, Sean. Good morning. I'm going to give you some insights on Praxair Distribution, Inc., which is the third leg of our vertically integrated supply system, that's integral part of our very impressive North American industrial gas business.
So we have an impressive national footprint to service our over 0.5 million customers. 15,000 of these customers represent about 75% of sales. And we have 400-plus dedicated sales folks that provide attention to these critical customers, to make sure we grow with these folks. We have over 1,000 packaged -- we have over 100 packaged fill plants, 5 tier 1 specialty gas plants, over 1,000 trucks making 3,000 deliveries a day. Most importantly, our 4,000 associates in PDI are committed to providing value to our customers day in, day out.
So who are these customers? First of all, they vary and they're very diverse. Manufacturing and met fab is our largest segment. An example is Boeing, who order us all their business on a national business, including their new site in Charleston, where they're going to make the Dreamliner. Our growth is driven by return to competitiveness in U.S. manufacturing. As you know, automotive annualized production is now over $16 million, we're seeing torquespace [ph] and aerospace and in transportation equipment. The energy infrastructure buildout due to shale gas, tight oil formation, especially in Dakota, Texas, are driving demand for tanks, trailers and associated equipment. Future construction of petchem projects in Texas and Louisiana will be strong drivers for us and we look forward to seeing those come to fruition. Labs and universities drive research where the U.S. has no peer, and this segment provides very nice growth for our specialty gas business. So overall, we have a nice $1.5 billion business with very strong cash flow, 17% net operating cash flow.
Let's talk a little bit about our growth drivers. Metal fabrication, anybody who cuts, joins or processes metals, represents 35% of our sales. We have a differentiated StarSolver Program, which has been the linchpin of our success with our customers. We typically attack about 85% of the cost stack of a welding shop, and can almost guarantee a 20% cost reduction to this cost stack. How do we do this? We have over 30 dedicated engineering specialists that drive around the country, and they're supported by our very qualified labs and other capabilities. I'll give you an example. Cameron in Williston, North Dakota, which is at the heart of the Bakken Shale, they make 500-barrel crude oil storage tanks and needed to dramatically expand capacity. This is in a market where welders are scarce and they have quality issues. We automated the equipment, we have customized the equipment to perform rolling, cutting, welding at different and difficult angles, increased capacity by 4x, but most importantly, without any headcount addition to the customer. They rewarded us with a long-term agreement and we've got numerous examples like this.
Specialty gases are growing at double digits, and there are some really nice, secular long-term prospects in this space. First of all, tighter EPA specs on emissions at driving, monitoring at lower levels and volumes. Engine turbines at power plants, down to chainsaws, lawnmowers and blowers require very frequent testing. There's increasing demand for natural gas standards at each junction, and we've got large processing facilities. We're also excited, we reported an acquisition named PortaGas, and we're excited about this offering as we grow this business. Daily calibration of confined space and personal safety monitors, typical end-vendors such as Draeger MFC and Industrial Scientific, and we have leading positions with these folks.
We measure gases like methane, carbon monoxide, chlorine and hydrogen sulfide. Toxic gases, at low levels, that ensure the safety and personal hygiene of our folks. PortaGas offering is unique, it has an ISO cylinder for global use, currently only packaged to do so. It's an industry-changing package, it's the first one that has a refillable cylinder, as opposed to a disposable cylinder, and that drives sustainability with our customers. We're growing this business really, really fast. I'm excited about it.
Finally, let's talk about improving our business. We've got really strong growth prospects. The industry consolidation and dynamics and our focus on pricing and productivity have helped us improve the performance of the business. First, let's talk about the industry. This business has been steadily consolidating, as you can see on the chart on the left, as more and more independents exit the business. There are still 700 independents out there and we work hard. We have been very proactive, have a dedicated and M&A team that's focused on this space. This team also has its own dedicated DD and integration teams that stays with the project, post-close, to ensure that we have the synergies in the first 12 months. This is critical for a good return to the project. We've closed 80 deals since 2001 and are starting to see the very strong benefits in our financial performance of the business. The results are increased density, and we of standardized standardize processes with this business we've brought on board. I'll take you to the middle side, which is Texas, as a great example of this. We have really strong air separation and bulk assets in Texas a few years ago, access to large customers and their vendors. We decided our packaged gas business was nonexistent. We scratch started and it supplemented with some key large acquisitions. Results have been really positive, and we've seen strong sales growth and very, very strong operating profit leverage, as you can see on the chart.
Now let me talk about continuous improvement at Praxair, and our packaged gas business has a lot of room to improve there. As you look at the bar graphs on the left, operating margin versus degree of fragmentation in the industry, you see improving performance with higher consolidation, sounds very logical. The bar to the left is also, very coincidentally, where most independents are and where we were in 2001. Through productivity, pricing and building density, we improved our operating performance by 600 basis points during this time period. We also expect, as we continue to build out our consolidation strategy and more pricing and more productivity, we could get to the levels of productivity and profitability that we see in our most -- best performing businesses in Mexico and Canada. And now the little guy with the 5-pound-a-day business is now going to hand it over to the 5-million-pound-a-day business, and Dan Yankowski and our global HYCO business.
Good morning. I'd like to talk with you about the global demand for hydrogen, how we see it around the world, talk about our hydrogen projects and then the U.S. petrochemical industry and the opportunity that we see for industrial gases.
The demand for hydrogen, globally, is still robust. We estimate that over the next 5 to 10 years, we'll need an additional 8 billion cubic feet of hydrogen globally. That's being driven by new refineries, expansions, upgrades and the use of heavy crudes. There's 2 other trends we see occurring. The first is the amount of diesel. Diesel consumption is growing 3% to 4% per year. And when you look at how diesel is produced, it's a cut out of the crude distillation unit. And if a refinery wants more diesel, they need to heavy up the crude coming in to the distillation column, and they also need to install hydrocracking technologies, which require large volumes of hydrogen. The second trend that we see is the adoption of low sulfur fuel regulations. The U.S. EPA is pushing for tier 3 mandates by 2017, that means that gasoline in the U.S. at 30 parts per million needs to go to 10 parts per million. What does that mean for hydrogen? It means that we'll need an additional 100 million to 200 million cubic feet per day of hydrogen in the refining industry in the U.S. alone. Look at China. China is about 350 ppm going to 10 [ph], by 2017. This, alone, will require about 1 billion to 1.5 billion cubic feet per day of hydrogen. So the demand for hydrogen is still very strong around the world.
There are some headwinds that we see, not in the demand side but on the sales side. The first is in Latin America. A lot of the projects, refining projects, upgrades, new refineries, have really been slow to develop. In some cases, we've been working on projects for 6 to 7 years, and those projects have -- just moving forward now or have been canceled. At the same time, a lot of the state-owned oil companies want to own their own hydrogen assets.
Looking at China. China, their refining capacity today is about 12 million barrels per day, growing to 16 million barrels by 2017. They'll need significant quantities of hydrogen in the refining processes. Due to the lack of natural gas, natural gas infrastructure and the high cost of LNG, most of that hydrogen will be produced through coal gasification. So we won't see a lot of SMR-based hydrogen projects, but we will be seeing a lot of gasifiers. And that is an opportunity for us to sell large volumes of oxygen to these gasification projects. You'll hear more about that from Anne Roby, later this afternoon.
Now, in the U.S. we see the demand continuing to grow. A lot of that is driven by the development of shale oil and shale gas, which has significantly improved the overall competitiveness of both the refining and petrochemical industries. The light crude have a moderating effect on the amount of hydrogen that's consumed in the refineries, but at the same time, couple that with low-cost natural gas and a lot of the refineries are very competitive. If you look at the chart in the upper right, you'll see that today and over the last 3 years, the U.S. refining industry has gone from a net importer to a net exporter of refined products. Today, we import less than 1 million barrels per day of gasoline and we export significant quantities of diesel. At the same time, because of the U.S. crack spreads and the profitability of the U.S. refining industry, they're operating at record utilization rates. These facts have offset the impact that low -- the light crude has had on the hydrogen demand, and we continue to see an increase in the demand for hydrogen around the U.S. refining industry.
The U.S. petrochemical industry, with the availability of low-cost natural gas and the natural gas liquids, has really gone through a revitalization. I'll talk about it in subsequent slides. But at the same time, if we take a look at the opportunities to supply industrial gases, hydrogen and syngas to the petrochemical industry, couple that with hydrogen and the hydrogen plants that we have in construction today, as Steve mentioned, we see a doubling of our business between now and 2017.
Let's talk about our hydrogen projects that we have in construction. I'm very happy to tell you that we've completed 3 world-scale hydrogen plants. The first one is in Port Arthur, Texas. It's up and running and producing hydrogen to supply the Valero refining company, new hydrocracker there. That plant is also coupled with our hydrogen pipeline system that travels from Louisiana into Texas, along with our storage cavern there. And we're able to supply not only Valero but all of the customers who are connected to that pipeline system with their increased volume of hydrogen.
The second plant that we recently completed is in Norco, Louisiana, again, for Valero, to supply their new hydrocracker with hydrogen. That plant is also up and running. It's connected to a new 50-mile pipeline that runs from Norco back to our Geismar, Louisiana facility, and along that way, we're making connections to most of the refiners so we can supply their increasing needs for hydrogen.
Lastly, in Paradip, India for IOCL, who's recently completed one of their world-scale refinery, we've completed a plant. It's ready to go. And once IOCL is complete, we'll be able to start that plant up and supply them with their hydrogen. This is really the first major hydrogen plant that's third-party supply to refinery in India, and it reinforces our ability to supply future companies in the future.
Let's talk about the U.S. petrochemical industry. As I mentioned, the refining industries have significantly improved our competitiveness. The U.S. is doing the same -- petrochemical industry is doing the same. We see that with low-cost natural gas and natural gas liquids and the discount or the spread between that and crude oil, that the industry itself is undergoing a renaissance. Three things that we see. The first is the return of C1 chemistry. That is picking natural gas and converting it to methanol and ammonia. Those chemicals require large volumes of hydrogen. And since this is gas, which is hydrogen and carbon monoxide, depending on the technology, you could require large volumes of oxygen to produce that syngas.
The second is the availability of low-cost ethane and, again, the disconnect of ethane from the crude oil price. What that's doing is fueling an investment in new ethylene plants in the U.S. Today, I think there's an estimated 10 new ethylene plants being announced in the U.S. Several of them are already in construction. There's not a lot of industrial gases in the production of ethylene. And ethylene is tough to transport because of the type of molecule it is. So they build downstream chemicals, ethylene oxide, ethylene glycol, polypropylene and polyethylene, all require large volumes of oxygen, nitrogen and hydrogen.
The third thing happening is because of the availability of natural gas and the disconnect from crude oil, there are several GTL plants in development today, and they're taking advantage of this disconnect. One GTL plant requires a large amount of oxygen equivalent to the total amount of oxygen that all major industrial gas companies produce today on the Gulf Coast. So with that, there's tremendous opportunities for industrial gases in petrochemicals. In a way, Praxair, with its ability to develop product line plants that we talked about earlier to engineer, design and construct plants, to leverage our infrastructure, will be well-positioned to grow both with the petrochemical and refining industries in the U.S., as well as internationally.
I'd like to turn over the mic to Matt White, our President of Praxair Canada.
Matthew J. White
Thank you, Dan, and good morning. What I'd like to talk about in Canada today is that -- and we have a high-value business in a developed market, but where we see some growth going forward is the tremendous amount of investment and development and the wealth of natural resources in that country. And those are really secular drivers for the long range.
So when you look at the first slide I have here, the upper left quadrant, you can see that we have the leading position in Canada, with 2012 sales of $1 billion. How do we get there? We've been in the country for over 100 years. We have a veteran team. We're one of only 2 industrial gas companies that has a fully integrated coast-to-coast supply network. We have very diverse end markets. And we have, in my opinion, the right density in the right geographic areas. And when you look at these geographics, I think it kind of gives that picture some clarity.
First, on the integrated supply. We have the traditional supply modes, on-site, merchant and packaged, but we also have a strong gas service network. You can see the healthcare services, oil well service business, we have [indiscernible] business, similar to NuCO2, nitrogen pumping and several other type gas-related services.
We have very diverse end markets. You can see the spread. Now manufacturing probably about 40%, similar to a John Panikar show, a lot of that met fab. A lot of that is automotive or aerospace-type firms.
But when you look at the rest of the pie pieces, you've got a combination of both cyclical and noncyclical industries. And if I were to show this graph 5 years from now, I think it'd look quite different. You'd see the energy, metals and, to some extent, chemicals pies growing at a larger pace, and they would become larger relative to the manufacturing and healthcare.
And that takes you down to the map at the bottom of this slide. And I think the first thing you can see with it is just the density and the spread of the dots. And these dots represent our production assets in atmosphere gases or CO2 plants, and the smaller dots represent some of our packaged and healthcare locations, fill plants and branches. And you can see we've got every province covered. And we have density not just in the highly populated areas, along the border in the major cities, but also in the Northern and the Western regions where a lot of the resource development is happening today.
And when you look at the Canadian economy, I think simplicity, you can break it down into really 2 almost subeconomies. The first is in the East, and that's represented in the lighter green, primarily the provinces of Ontario, Québec and the Atlantic provinces on the East Coast. And really, that's a manufacturing and a mining-type economy. Now as Steve mentioned, we're seeing a little bit of headwinds in Canada today, and part of that is the manufacturing. The southern part of Ontario and Québec, as you know, decades ago, they achieved currency that was an opportunity for some labor arbitrage to set up manufacturing, primarily for U.S. companies. And now with the currency of parity, it's creating some headwinds for our customer base, but that does still create opportunities for us as well.
First, with our supply network in the U.S. and Mexico, where some of these operations are moving to, we're well-positioned to retain that business. But second and more importantly, as Scott Kaltrider mentioned and Sean Durbin, we sell productivity solutions to our customer. We've had a lot of wins, similar to John mentioned, in Québec and especially the Greater Toronto Area with our productivity solutions to save our customers' money. In the northern parts, in the eastern provinces, it's mostly mining and metals, and I'll talk to that in the next slide in more detail.
And moving out West in the dark green is more of a natural resource play, and that's really mostly British Columbia and even more so, Alberta. And I'll talk about that in the next slide as well, but that's a big hydrocarbon opportunity that I'm sure a lot of you read about in the newspapers over the last several years.
So I think to sum up on this, we have a viable franchise, and while it's a developed market, there's still some great opportunities in the natural resource side.
So now I've broken this slide really into the 2 main secular drivers that we see in natural resources, metals and mining on the left and oil and gas on the right. Now the Industrial Info Resources recently came out with a figure that said if you add up in Canada, all of the projects that have been announced, that are in process equates to $0.5 trillion. Now I don't think all these projects are going to happen. Some may be deferred, some may be canceled. But when you look at $0.5 trillion in an economy that has a population and GDP roughly equal to the State of California, that's a significant amount of investment. And about 1/3 of that or $170 billion is really around the metals and mining and, more specifically, development of iron, copper, nickel, gold and, to some extent, lithium, rare earths, uranium.
So there are tremendous opportunities for industrial gas company like Praxair with that development. The first is our basic atmospheric gases that we supply the steel production, oxygen, nitrogen, argon. A recent example, T900 [ph], a 900-ton-a-day plant that we had recently stalled in Essar Algoma in Sault Ste. Marie.
In addition, we supply oxygen into smelting and leaching operations, and we've seen a lot of opportunities there. Some recent ones that will be starting up here soon is Vale Newfoundland for their nickel upgrader, as well as a couple of gold mines that we recently won in northern Québec and northern Ontario.
And also, we see a lot of water treatment. With these remote mines, they need to use processing water, and they've got to clean that water for environmental purposes. So we get oxygen and CO2 pH, and we're well-positioned to capture that because it requires liquid product to be transported. You need to have production plants in those remote areas.
Now looking at the oil and gas opportunities, that first one, the oil sands, and this is the one I think that gets a lot of air time and a lot of newsprint. And probably the most significant piece of that is the Keystone XL Pipeline. Now from my perspective, the Keystone is just one of many projects that are underway to monetize the assets and the hydrocarbons in Alberta and B.C. And, frankly, I'm sort of agnostic whether it happens or not because when you look across, what are the projects that are currently underway to get those assets and get money out of them? Enbridge has a Northern Gateway Project they're looking to get those hydrocarbons to the British Columbia coasts for export to Asia. TransCanada has an East Canada Pipeline to send it to the refining assets in Eastern Canada to displace imported crude. Enbridge also has another South Pipeline expansion they're looking to get product in the U.S.
The significant rail activity has been growing to move the diluted bitumen out to Northern -- North American refineries. And even the province of Alberta is investing, and I'll talk about this in the next slide, to just further refine the hydrocarbons and distribute the higher value product within the local markets.
Exploration and frac-ing continues to be a big opportunity. 6, 7 years ago, it was all about dry gas. Now, today, it's all about the barrel, it's all about liquids. So the majority of the frac-ing we do today is for either oil or for natural gas liquids. But as the LNG and, to some extent, GTL infrastructure build out, dry gas will be further growth even on top of this. When you look at the LNG investments that are happening today, the big part is Kitimat, you probably hear about, in British Columbia and Prince Rupert. And it's an attractive opportunity because you have significant dry gas reserves. You have a very friendly investment economy in Canada, very stable and a wealth of resources, which is not -- you can't say that around some of the other countries in the world that have this wealth of resources. And it's not just in Canada. The private companies have opportunities to invest the big integrated majors like Chevron, but also, some of the state-owned oil companies also have investment opportunities.
So I think from that perspective, LNG is a good growth opportunity and, to some extent, gas to liquids because you look at northern Alberta, their long dry gas and their short liquids and the liquids primarily used as a diluent to transport bitumen. So the opportunities for us there, clearly, liquid nitrogen, liquid CO2 to support the enhanced oil recovery in the fracturing market.
In packaged gases and hardgoods really can apply to all of these infrastructure development. Now if you're going to build 1,000 kilometers of pipe, you need a lot of welding. If you're going to build a very large mod in northern Edmonton that needs to go up to Fort McMurray onto the oil fields, you need a lot of wells. So as this infrastructure is built out, that's going to be positioned to benefit for our packaged gas business.
And the last point here is oxygen, hydrogen for bitumen upgrading, and I'll talk more in detail on a specific project on the next slide related to this. So when you add up these sort of secular drivers that are really long-range, mining had a dip here recently, but the customers we talk to are looking out 10-, 20-year-type horizons. Plus, the organic growth that we expect in Canada, we can see ourselves at $300 million growth over the next 5 years.
So in this final slide, I just want to highlight a project that we had announced earlier this year toward the end of the first quarter, and it's the North West Redwater partnership. So what is NWR? It's a 50-50 joint venture between North West Upgrading and Canadian Natural Resources Ltd. And how did we win this project? To some extent, it was a bit of a labor love. It was a multi-year process, but we have veterans in Alberta with detailed industry knowledge and gas knowledge. And this has started several years ago when NW is more of a small startup, but it's making these connections at this level that can get you to grow with your customer base and win some of these. So we have a lot of these in the hop or all around the country every day, and some can become a project of this nature.
They're looking to invest $6 billion for startup in 2016, and they'll be located in the Greater Edmonton Area, as you can see in the map. And north of that will be Fort McMurray, where they would either mine out there or SAGD the -- get the bitumen out of the ground and pipe it down to this refinery. The interesting thing about this North West Redwater partnership is it's really a tolling operation. They don't take risk on the barrel, they don't take risk on the spreads and the risk to build the plant and to operate the plant. They just need to process the barrels, with some upside that they will achieve a goal over certain production targets. So really, it's the Albertan government at 75% that they're providing the bitumen that they receive as a royalty, and then they're taking ownership throughout the barrel to the finished product. So for Alberta, it's an opportunity to diversify their revenue source. They have a significant exposure to Western Canadian select crude, as well as bitumen pricing. So by doing this, they can be more diversified and enjoy the value down the hydrocarbon chain.
On the lower right here, you can see a very, very basic schematic of what this project is. But this first phase, and it's one phase of a potential 3 down the road, will be 75,000 barrels a day diluted bitumen. They would basically vacuum out the diluent that they would refine but then take that heavy bitumen residue that they would put through a hydrocracker, similar to what Dan was talking about, to basically break that longer chain down. And that will require the gasifier that needs our oxygen. Now an interesting thing about the gasifier is they're actually going to take the CO2 offtake and be a clean CO2 stream, they'll compress and they're piping a couple hundred kilometers through one of their sister companies for an EOR application in Alberta. So really, this is almost like a closed-loop carbon system, and it will be developing local bitumen into high-value finished products like ultra-low sulfur diesel, the vacuum gas oil, naphtha and other lighters that can be sold in that area. So it's something we're excited about. And I think it creates an opportunity not just with this or potential expansions, but if this organization builds out and you get petchem or other type upgrading opportunities, they'll continue to grow.
So that concludes Canada. Next up, I'd like to introduce Antonio Cesar, who's President of Praxair Mexico.
Antonio Cesar Miranda
Thank you, Matt. Good morning. [indiscernible] share with you a little bit about the moment of Mexico, why we like it and how Praxair will play in this scenario going forward. Experts have said that Mexico is one of the most promising place to be over the next few years from a business perspective, and here, I list 4 reasons for that. First of all, as you all know, Mexico is one of the most open economies with 44 freight trade agreements with all countries all over the world, and this has been the bright of Mexico since it opens a great market to keep the focus on exports, which have been one of the Mexican growth platforms.
The second reason is competitiveness. In late '90s, beginning of 2000s, Mexico could not match China's extremely low wage rates, as illustrated by the bar chart down on the left. Manufacturers located the factories in China rather than Mexico. It was a crisis that Mexico did not like to go to waste. They did the homework. And along with the strengthening democratic institutions, the government began to focus on improving infrastructure and alleviating unnecessary regulatory burdens.
Meanwhile, Chinese wages have skyrocketed, and transportation cost is affected by higher fuel prices, began to rise extremely fast. So in this direction, total land and manufacturing costs for goods made in Mexico are currently lower than in China. In this sense, manufacturers are very optimistic about the future of Mexico. And these positive trends and developments have been attracting many foreign direct investments, as represented by the pie chart, that have surpassed the $100 billion of investments, being 46% captured by the manufacturing sector. Within the manufacturing sector, automotive companies have been leading the way with $10 billion of investments. So simply put, Mexico has what manufacturers are looking for, low labor costs, educated and skilled workforce, reduced shipping time and cost and access to markets.
Another very important sector for Mexico is the energy. Despite the oil production decline, Mexico is still among the 10 major global oil producers, with the sixth largest estimated shale gas reserves and a large amount of proven oil and gas in conventional reserves. So expectations today are on the energy reform approval. It seems to be a consensus that we take Mexico to [indiscernible] in terms of attractiveness going forward. So Mexico is extremely well-positioned to compete globally and be a very attractive market going forward.
And how Praxair will play in this promising and exciting scenario and will succeed in the Mexico industrial gas market? Firstly, we have a leading position in this very attractive market with mainly 2 competitors, which means we hold a big share of market, enjoy an OP margin above the average of the North American segment, being the preferred partner among key customers that is confirmed by our win rates. Only to give an idea, over the last 5 years, we've been able to win about 80% of the on-site projects. And in merchant, new business acquisition rates are in the range of 70%. So we are a $700 million company, being the most integrated player in the region, with a very dense packaged business that accounts for 40% of total sales, with more than 90 sales branches spread out all over the country, which give us a tremendous advantage in terms of capturing new businesses. We hold a complete portfolio of atmospheric, process and specialty gases and a balanced mix of hardgoods and safety equipment. This portfolio also includes maintenance services through injecting nitrogen and CO2 for the oil market, and we also lead the merchant market with the very solid medium- and long-term contracts, offering our customers a wide variety of technologies and technical support. So we are very well-positioned to succeed in this market that is a valuable franchise for Praxair in Latin America.
And to better illustrate what I'm talking here, let me share with you 2 key businesses that we've been successfully growing over the last few years, and we'll surely keep the pace going forward. Firstly, the packaged business that has grown at 10% rate since 2009. And on packaged, our major focus has been on 3 segments. Firstly, the metal fabrication. Metal fabrication, we truly excelled at developing technology applications like argon and argon blends for welding, heat treating and gases for laser cutting. In this particular application, we've been closely working with the OEMs. Specialty gases. Specialty gases has been a very important growth platform for Praxair in Mexico, and it is important to mention that we are there in Mexico Center of Excellence of Praxair worldwide in terms of specialty gases manufacturing and techno knowledge. And this is a market where we clearly differentiate ourselves and have grown above our fair share of market. Our -- the main markets of specialty gases include energy, environmental and food and beverage.
And finally, the Retail segment. We have made progress in terms of improving our market coverage through low-cost retail stores, and our plan includes opening 16 new retail stores over the next 2 years.
Even with respect to package, let me tell you a good story about the microbulk conversion program. It basically consists of a business that converts larger volume of gases and cylinders into small tanks, and the science here is the business model. We appropriate tanks, these small bulk delivery vehicles, the [indiscernible] installations and the logistics systems. It is very attractive to customers since it is safer. Once customers know longer have on sealers and cheaper because it reduces the probability of losing synergies, it allows the better control of inventories and dramatically reduce the number of invoices that a customer handles on a day-to-day routine. And of course, for Praxair, it lowers -- it means lower cost to serve, with businesses 5 to multi-year contracts with price escalation from and the rental fee for equipment. So this is a key business that we've been growing 10% of growth rate over the last 3 years and very important for our growth platform looking forward.
Another very important sector for us in Mexico in energy. And I guess you all know that PEMEX has been struggling a little bit with the oil production decline over the last 10 years going down from 3.5 million barrels per day to 2.5 million, with the existing wells requiring more stimulation, more maintenance services, more recovery techniques, while new drillings requiring knowhow, new technologies and huge investments. So this is the reason, this is the main reason why the energy reform is eagerly awaited. But luckily, the struggles of PEMEX have represented great opportunities for Praxair to benefit even further in this market with our existing products and services in 2 different fronts.
Firstly, the EOR, the enhanced oil recovery, where we inject about 6,000 tons a day of nitrogen into Samaria oil reserve. And the second is the Nitropet, which is our bulk type of business, where we inject nitrogen or CO2 basically for maintenance, clean and fracking. Far from being a traditional sale of gases, Nitropet is an oil service company that, along with the technology applications and technical support, reliability and safety have been the name of the game and our competitive advantage. And this is the reason Praxair has invested in air separation units, pump units, trailers and storage receivers. But the key aspect on this market has been the more than 200 people who have gone this business at 20% growth rate.
So I move to the Q&A session and invite Steve to lead the session. Thank you.
Stephen F. Angel
So Antonio, you didn't want to stay up here and handle the questions? Okay. Well, if we get one on Mexico, I'll let you handle it, or Brazil. Questions? Just pick one.
I'm Dave Begleiter, Deutsche Bank. You mentioned annualized capacity. What do you believe is the dollar amount of the annualized capacity? I mean, on what margin will that capacity come on stream going forward, incremental margin on that capacity?
Stephen F. Angel
Yes. The question refers to underutilized capacity and what's the value of that, which I have not calculated in dollar terms. But I'll just give you some percentages around the world. And you'll see that it's significant. So if you look at Europe, our capacity utilization. And first of all, these numbers can move around by products, so I'm cautioned by my people not to speak too much in absolutes because in the U.S., argon is very tight, whereas Lin/Lox is not. But if you look at Europe, Europe basically has about 60% capacity utilization on the merchant liquid side. Packaged gases are about 2/3 of what they were prerecession. So you've had a lot of volume as we lap the marketplace. And we've got a long ways to go before we would ever, ever need to add another merchant liquid plant in Europe. Again, 60% capacity utilization. If you look at Brazil, the numbers would be, I think, in the high-70s, low-80 s, so we certainly have some underutilized capacity there. It could vary depending on what part of Brazil you're talking about. But we have a ways to go there before we would need to add merchant liquid. If you look at the U.S., Lin/Lox, you saw it on the charts, mid-70s, so certainly, we have some room to run there. Argon, much tighter; hydrogen, much tighter; helium, extremely tight. And if you go to China, I think capacity utilization, a little higher in China than it would be in the rest of the world. But that's kind of how you would rack and stack it, and you know you can use your models to kind of come up with a figure. But clearly, we've got -- we have capacity around the world, largely in the merchant liquid. And as those volumes return, we expect the margins to fall through pretty heavily. We just need volume. So once -- excuse me. Kevin, go right there. Tammie.
Kevin W. McCarthy - BofA Merrill Lynch, Research Division
Kevin McCarthy, Bank of America Merrill Lynch. Stephen, years past, I think you had a sales growth goal of 8% to 12% and EPS was 12% to 18%, so about a 50% multiplier there. Today 8% prospective sales and double digits for earnings. And so my question is, has there been any change in the relationship between sales growth and your anticipated EPS growth when you think about operating leverage or contribution margin? And any changes in taxes or leverage?
Stephen F. Angel
Not, not really. I think taxes, no. Leverage from share count reduction, no, no change. And I think clearly, if you get strong volume growth, you can get more leverage off that strong volume growth than if the growth is weaker. So I mean, if I'm looking at something in the order of 4% or 5%, 6%, then I get some leverage on that. 1 or 2 points I think is a pretty good job. If you get 9% sales growth, I think we can get 12% OP growth, for example. So nothing has really changed in my mind around the leverage. It's just stronger volume, I think you've got more to leverage than you do with weaker volumes, simple as that. P.J., go ahead.
P. J. Juvekar - Citigroup Inc, Research Division
P.J. Juvekar from Citi. Steve, you talked about your CapEx being $1.8 billion.
Stephen F. Angel
Spend, annual spend rate.
P. J. Juvekar - Citigroup Inc, Research Division
Annual spend rate?
Stephen F. Angel
P. J. Juvekar - Citigroup Inc, Research Division
And the project backlog, it seemed that it peaked in 2012. So going forward, do you think this backlog will be steady or will it begin to decline?
Stephen F. Angel
I think it's going to drift down to around 2, and the reason I say that is -- I mean, I'm already seeing a lot of the proposals we've had out there just take a long time to close. Not losing anything, but it takes a long time for these projects to close. And I think people around the world are doing what they typically do. When it's kind of uncertain times, they're just reluctant to pull the trigger. Now they still do all the business development work. Some people stay busy, but you just watch that it takes a longer time for things to close. And that's what I'm seeing today. So $1.8 billion, again, is a pretty healthy rate going forward. It will support 3% to 4% project growth. If I'm wrong and the economy has picked up around the world and project activity moves off the back burner on to the front burner, then we're ready to capitalize on them. So I'm not -- you're going to accuse me of being a little conservative in that front, maybe. But I'd like to see the whites of the eyes before I start to declare victory. And when you look at the mix of the project backlog, probably I'd say 40% or so is tied to things that really aren't tied to industrial production capacity growth like hydrogen and gasification, and decaps [ph] and things like that. So you've always got that going for you. But the rest of the backlog, the rest of project activity is industrial production growth. And if it picks up and moves forward, we'll be ready -- we'll be prepared. But that's just not our operating assumption today.
P. J. Juvekar - Citigroup Inc, Research Division
And just a quick question on GTL. I think you just talked about I would need multiples of...
Stephen F. Angel
Yes, it's big.
P. J. Juvekar - Citigroup Inc, Research Division
What is the viability of these plants? And then if you invest in it, what is the risk for you in building such a large plant?
Stephen F. Angel
Well, I mean the people who are working on these projects, [indiscernible] so example, seems very committed to doing this. They're going through their process, which is very, very extensive -- permitting, front-end loading, all those things for a project of a -- if the number serve me right, something in the order of $18 billion, $20 billion. These are massive projects, right, so they move very slowly. I think if it went forward and started up, it's probably closer to 2019 before it happens. It's out of this 5-year time horizon that we're talking about. From an execution standpoint, they're just going to be multiples of plants that we built before or are very close to the size we built before. So that part doesn't concern me all that much. I just think it's going to take a long time for their project to move forward. And again, it's out of this 5-year time horizon in terms of the effect it might have. Now one consequence of that, which we won't have to deal with for quite some time, is that the plants of that size can obviously put a lot of argon into the marketplace. And that's not something I would be anxious to see. Again, it's a long time before I would have to deal with that. But we kind of like tight supply demand markets, and that would have a potential to be somewhat disruptive on the argon standpoint, which is something we'll have to deal with if we got closer to that. Go ahead right there. And I'm going to start moving to the back and pick out people.
Just to piggyback on that. Can you clarify what your economic growth assumptions are embedded in the 5-year outlook both in terms of GDP and in terms of global industrial production growth?
Stephen F. Angel
Well, we think of in terms of industrial production growth, I said roughly 3%. And I'm embedding some of the application development work in that contribution from that. And you can call that 1 point or so. So roughly kind of 2% is what we're working off of in this time horizon. Go right here, Tammie?
John Roberts - UBS Investment Bank, Research Division
John Roberts, UBS. Steve, on Slide 18 that shows U.S. merchant pricing dynamics versus capacity utilization, it looks like there's a 12 to 24-month lag. Roughly pricing was strong during the depth of the financial crisis, and I assume that relates to the contract term. So given the drop in utilization recently, are we headed in the next 12 to 24 months on a continually weakening U.S. merchant pricing?
Stephen F. Angel
No. No, we're not going to be seeing weakening merchant liquid pricing, or Scott Kaltrider would be fired. But no, what happened in the '09 timeframe is you got a lot of mix things taking place here, where you got various customers, and volumes drop. And that could kind of all go into the weighted average price calculation. So there's some mix effect there. But the point of that chart really was to say that even when base volumes drop 12% worldwide, we were able to hold price and even climbed up on pricing a little bit. And so what we really wanted to leave you with on that chart was that capacity utilization is relatively low today as capacity tightens up ever so gradually. And we see higher capacity utilization that the pricing dynamics become more favorable. So if you go back and look at where we were historically, when pricing gets into the mid-80s -- excuse me, when capacity utilization gets into the mid-80s, pricing dynamics certainly improve. So I don't know how long it's going to take to get back there. I mean, we're certainly not going to be adding any merchant liquid capacity. So as volumes progress in the U.S., it will tighten up. And my only point is pricing will be better. So I said 1% to 2% earlier, volume growth. Some volume growth will put us closer to kind of about 2% across-the-board, if it -- volumes. If they were to flatten out, we get no growth. I think we'd still be able to deliver 1%, for example. So this gentleman right here.
Edward H. Yang - Oppenheimer & Co. Inc., Research Division
Ed Yang, Oppenheimer. You mentioned the low utilization rates across various geographies, and you expect volume growth to grow about 2%, 3% a year. So it would take a while to soak up and have that utilization rate improve. Does this mean that you expect to see some industry or capacity rationalization or consolidation or possibly asset swaps. It's been many years since we've seen that? And on the flip side, would you expect to see competitors move out of their traditional geographies and try to edge out into new markets?
Stephen F. Angel
Well I think -- just taking your last question first. I think -- look, an early key, Linde, they're a major global corporation. They're already in just about every place they were in. So I don't really expect to see anything different in that regard. And as far as volumes growth, it won't happen formally across the globe. You'll have certain places that will grow faster. We're basically assuming no growth in Continental Europe going forward, so that kind of weighs down on our forecast. And based on that, we may be looking at 60% capacity utilization 2 or 3 years from now. But other parts of the world, we'll grow a little faster, and I think the pricing dynamic that I talked about will take hold. Back to your question on asset swaps, we're already doing some things like this today in a place like Europe. So it's not lost on the industry at large that there is some value. And kind of looking at our combined assets, not our individual assets and I'll just give an example. We've got a plant here, and across the street there's a competitor plant there, and we're both 50% loaded. And then you go 100 miles or 200 miles away, there's a pair of plants and they're both the same situation, each about 50% loaded. Well, it may make sense, and it has made sense, for us to idle our plant in the first region, just take product out of your plant, and then in the second region I talk about, do the same, do the reverse. So there are things that we constantly look at, our people constantly look at on a localized level, to really better utilize our assets, not only our assets, but in some cases competitors assets. And I would say there's much more of a willingness today in Europe to do that than there was some years ago. But this kind of economy makes people do things like that. As far as country swaps, the best example I can think of is when we swapped our Turkey business, no pun intended, but our business in Turkey for Linde's business in Mexico. And that was the swap we made -- I forgot how many years ago -- 2007 is when we consummated that deal. And I think you can tell from Antonio's presentation, that's worked out exceptionally well for us. And we continue to make those phone calls to our competitors and see what kind of discussions we can have about swaps like that, that can make a lot of sense really for both parties. Let's go to the back of the room here, Stacey?
Robert A. Koort - Goldman Sachs Group Inc., Research Division
Bob Koort at Goldman. Steve, I think Sean mentioned 500 or 5% productivity every year, which is pretty darn remarkable. So I'm wondering what eats away at that productivity? How do we see it from the outside? And then I think most observers would agree you guys run the most efficient gases company. And I was a little surprised, and I'm sure this will sound to picky, but over 20 years, your margin delta to the competitors only rose 200 basis points. Is there...
Stephen F. Angel
You're picky, Bob. Go ahead.
Robert A. Koort - Goldman Sachs Group Inc., Research Division
But you're so good. There must be a reason why it would have been more. Can it be more?
Stephen F. Angel
Can it be more? All right. Back to your first question, what eats it away? We have inflation. We do give salary increases in our company. You might think we don't, but we do give salary increases. We like to do that. We like to pay bonuses. We have -- we try to be very fair with the benefits package that we offer. So that's part of the inflation that's kind of built in every year. And electricity, of course. It's again largest part, outside of people, largest part of our cost stack is in the energy side. 80% of that is electricity, and I see power rates go up every year. So that's part of the inflation that we see. There's other inflation [indiscernible] built into that. That's why you go back to that example that I gave you, where if you have $100 inflation, you really need $200 between pricing and productivity to get the operating leverage that we want on that revenue. So another way to think about it, if price is equal to inflation and it never works this simply, then the productivity that I get, we put in our back pocket. As far as expanding margins going forward, obviously, continuing to build density. It's been a very successful strategy for us, but continuing to build density in every geography we're in enables us to do new projects at even higher returns because we can leverage the infrastructure that we already have. And obviously, all projects that look like that or sound like that, we're 100% focused on. Because we recognize it's not only good growth, it's good returns on those type of projects. So I'd say that, plus the whole focus on productivity. Pricing, we talked about that enables us to continue to grow operating profit, percent of sales, and I said 30 to 50 basis points per year. That was true 3 years ago. That's exactly what we did. We're better than that. And I see no reason why we can't continue that. Over here. Don?
Donald Carson - Susquehanna Financial Group, LLLP, Research Division
Yes. Don Carson, Susquehanna. You've identified hydrogen is a strong growth opportunity. But at $4 natural gas you're lucky to get 15%, 16% margin. So how do you keep hydrogen from being margin dilutive? Do you have more and more projects where you don't take title to the energy and boost your returns that way?
No I think, obviously, I'm factoring hydrogen to the mix. Low-cost natural gas means the operating margin as a percent of sales is actually higher than it would be if you had high priced natural gas. I'd say if natural gas went to $9 or something in the United States, that would have a fairly dilutive effect on our operating margin. But it's kind of all in the mix. The way I think about it, we've got all kinds of projects starting up all over the world. And when I say 30 to 50-basis-point expansion, I am incorporating in the fact that our hydrogen business is growing. In fact, many of the projects that Dan showed today have started up this year. The one with IOCL I think is a straight facility fee. We're not buying the feedstock. The other 2 with Valero, we are buying the feedstock. So I've already factored in a lot of the hydrogen impact, or it's already factored in from the projects we started up. And that's factored in my thinking about 30 to 50 basis points improvement going forward. Okay.
So I understand that was the last question I can take at this point? Oh, we were seeing volume perspective? Okay? So this will be the last question, then we'll go. We don't want to be late for our speed dating. I got to look around the world, I think I've given you a little bit of a view into what we're seeing in Europe. A lot of people talk about a turnaround. I have not seen that. I think it's stabilizing. But certainly, there's nothing that I can point to, to say things are going to get better in Europe. And I've said that our outlook is that volumes basically do not improve going forward.
Come around. Let's go to South America. South America just had a couple of recoveries from a couple of recessions. The last recovery really took place in earnest in March. We reported, I think, pretty good results in the second quarter. You saw that. Volumes had come back, pretty much broadly speaking. Automotive, it was performing better and so forth. And that's kind of continue to hold. I don't see -- I haven't seen any change from what we really talked about with respect to South America.
Going back to the July, again the volumes continued to hold better than I would say the headline news. If I were to get a surprise in the next 6 months or so in Brazil, my belief, maybe I'm a little conservative here, but I believe it would be more on the negative side than it would be on the positive side. But right now, the business is doing quite well. No real change from what we reported in July. In fact, I'll just -- you could just comment [ph] across the top.
We're not seeing any deterioration, further deterioration from what we reported in July on the earnings call. Likewise, we're not seeing any resurgence or step up in volume activity anywhere.
If I look at Asia, even back when people were really concerned about China, not that long ago, our volumes are holding up a pretty well. And we think they'll continue to do well. We're kind of calling it mid-teens kind of growth going forward. The base volume activity is okay. It's nothing great, but it's fine. We're seeing our on-site business that serves the steel industry actually is pretty strong. And I think that would correlate well with what you've kind of been reading about and hearing about. So I have no concern about China volumes. I think they're going to be fine going forward.
If I come back to North America, and I observe we have a lot of stake in North America. Matt talked about Canada. The manufacturing base has grown a little weaker. Currency hurts, and they have some other institutional kind of problems that's kind of coercing a move more South, and we're in a good place to pick that up. But I would say right now, not much change again from July.
And the U.S. is going sideways. That's the comment that I've made before, and that's still what I feel today. Now if you look at our hydrogen volumes the next time you see them, they're going to be up, and they're up really because of project activity.
And our volumes to the chemical industry are holding in pretty well. Again, pretty flat coming out of the second quarter, but it was -- it had stepped up in the second quarter, and that's just kind of continuing. And obviously, our chemical customers benefit from lower cost feedstock, a little bit of a residential housing improvement. Many of you guys here in the audience are closer to that than I am, but it's just kind of holding.
The same thing true in the steel industry. They've been doing quite well. You heard 16 million housing starts, infrastructure build out, manufacturing doing okay. So the steel industry has had volume, and we've benefited from that. But again, from the standpoint that I'm seeing it, improved, the answer is no. It's just kind of going sideways. Look at the merchant liquid volumes, the same thing. Packaged gas volumes, John talked about that a little bit. We've got a very good future. But hardgoods volumes are negative year-over-year. And I think you'll hear that from other people that participate in the industry. Gases are positive. Again, everything is just kind of status quo, flat moving sideways.
Okay? Lunchtime? Thank you.
Scott E. Telesz
Hello. Okay, is this on? Everybody please take your seat. I'm about to share Praxair's secret formula, and you're going to miss out if you're yakking on smartphones. Okay. My name is Scott Telesz. I have responsibilities that include Europe, Asia, Surface Technologies, sourcing and strategic planning. And today, I have the responsibility of kicking off the afternoon session. So I drew the short straw of that highly-coveted after-lunch time slot. And I'm addressing a room full of analysts about the topic of culture, so pretty soft and fuzzy. So Kelcey, all we need to do is raise the temperature in the room about 5 degrees and really set the tone for a snooze fest.
So now culture is an admittedly a pretty soft topic, and it's all about how a company behaves, the priorities it sets and the processes it goes through, the type of people it hires and promotes. And it's also a relative topic. It's relative to other companies, other organizations. And so some of my comments are going to be anecdotal. Some of them are going to be relative to other places I've worked, so either my recent 3 years with SABIC, a Saudi chemical giant or my decade with GE or my several years consulting with McKinsey to their capital-intensive manufacturing businesses.
But maybe, where to start is to say, "Why should you even care?" Well, while I'm not going to share any numbers with you, there's nothing you're going to be able to put into your fancy valuation models. Why should you even care about our culture? But I'd tell you, we get 3 questions pretty steadily from the investment community. Can you maintain pricing power? Will your disciplined capital allocation continue, or is the risk return profile changing? And how on earth do you drive productivity and a profit improvement from that at a rate twice that of any industry peer? And we would contend that all 3 of those questions are about culture. And we'd say that we have a distinctive culture at Praxair. It helps drive the outperformance versus peers. And as Steve alluded to earlier, probably the most difficult thing to replicate.
So naturally, like any of you would have done if you were analyzing culture of a company, the first thing that you would do is you would compare 2 farm animals. And what you see on the left is a show horse, and what you see on the right is a picture of a workhorse. And what I'm here to tell you is I have never seen a company culture more aligned with a workhorse than every enterprise I've seen in Praxair.
So let me explain this a bit more. So a show horse, it's about keeping your head up, speaking about your results, getting noticed. A workhorse is about -- it's easier to get your job done with your head down. Let the results speak for themselves. A show horse worries about the optics toward headquarters. What's headquarters going to think? Maybe headquarters needs to make that decision. A workhorse just cares about what works with the field P&L. A show horse leans on commercial savvy. A workhorse is all about operations. A show horse is a risky investor. A workhorse is a conservative investor. A show horse spends time on big, bold, futuristic conceptual things. A workhorse, well, they're grounded in the details of today. A show horse always chases new revenue opportunities, new fields of growth. Because you know what, the grass is always greener over there on the other side of the fence. A workhorse, well, a workhorse is just happy with the field they have, and they want to squeeze the existing assets. A show horse is all about words, about what may happen in the future. And a workhorse is numbers, about what did happen today.
And what we would contend is in the industrial gas industry because its different in other industries, you want to have a workhorse culture because that's what drives the best results. So a little bit more on priorities and processes. I have 4 thoughts on priorities, safety first, and Steve touched on that earlier in the day. Safety is the #1 priority of Praxair, and there's no faster way to get fired than violate safety procedures and put yourself or your coworker at harm. Now we have outstanding safety results across the industry, but we don't have 0 accidents. And we can always get better. And what sometimes people miss about safety is the only way you get great results is you have an operational-disciplined mindset, and you have a continuous-improvement culture. And when you have those 2 things, that enables you to drive excellence and so many other aspects of your business. It has positive side effects. So all the things that Sean Durbin talked about in terms of productivity, that's actually a byproduct of a safety-driven culture. So that's thought 1.
Thought 2 is we favor profit over size. This industry generates a lot of cash. There's a lot of investment opportunity. You can chase growth all day long. You can throw capital and growth all day long. But if your goal is highest operating profit and highest return on capital and not to be the biggest in the world, that's actually a much harder path to go down. And you'll hear later from Anne Roby on that about how that's our strategy in China.
Third is we have a very stable consistent strategy. Sure, we do strategic reviews of all of our businesses every year, and we grind through. But candidly, the outcome is pretty boring. It's a very minor Porsche correction on a couple of areas. Most companies that I've seen, and they're somehow not happy with where they are. So there's a more frequent change if they, let's say, in the commodity space, well, they envy the world of specialties. If they're not in a certain geography or pipeline, well, they better get into it. If they are running a volume play, well, you got to shift the price because they're unhappy with the results. And so we better change strategy, then the future will be brighter. And with every strategy change, of course, comes the organization change. We're not that way at Praxair. We are just fine with stay on course, no distractions, more next year.
Fourth thought is we clearly favor best practical execution over the big bold innovative. Now I'm not saying we're not innovative. We win innovation awards every year around the globe. But our approach in regular barriers of going to a more detail is a little bit of basic research, a medium amount of technology development, a lot of in-the-field application deployment with customers, and we think that's the best formula for growth in our industry. And we're proud of the fact that our R&D spend, frankly, is on the low side.
We're fine with no Hail Mary passes. We're not going to worry about carbon capture and sequestration or advanced transportation fueling technology. We're okay if somebody else spends that money, and if it pans out, great. We'll jump into the party later, and we'll win through execution.
Now some thoughts on some of our processes. We have very strict monthly operating rhythms. So when I came to Praxair, from SABIC and GE, I thought, "Okay, first quarter close, here we go." Two-minute and hurry up offense, right? That's what happens at every big company. 13th week of every quarter is the big push. Get the orders out the door, double and triple overtime. Well, 3.5 years later, I'm still waiting for that quarter-end push. Now I'm not saying we don't hustle to hit numbers. We take our results commitments as seriously as anybody. But partly the nature of our business, minimal physical inventory, minimum take-or-pay agreements, the business runs steady. Any process industry runs best when you run it steady. So a Friday, the end of a quarter at Praxair is like any other Friday. We are a very hands-on business, incredibly hands-on. Whether it's a large or small capital project in China, whether it's a productivity review of a packaged gas filling station in Italy, whether it's the contractual language around a Houston pipeline customer, the pricing discipline, and Scott Kaltrider touched on it earlier today, you don't get pricing discipline by big sweeping statements from the headquarters. You get pricing discipline by thousands of salespeople drilling into the details customer-by-customer because you make money in this business 1 nickel at a time.
We are a very efficient communicating organization. I have yet to attend a staff meeting at Praxair, and you know what they look like. Most big companies have them right. Every quarter, a couple of times a year, everybody gets together, and you sort of talk about what's going on or what everybody's doing. And you sort of feel good about things. We don't do them. We don't waste time with it. If you have something decide, you have an investment to make or a part of the business to drill in, you send the information out, you get together. If anybody have any questions about the recommendation, you go through things quickly. More often than not, the meeting ends early. We just don't waste a lot of time talking to one another.
And accountability and the restrictions of the cost control, they are everybody's responsibility because all companies go through periods of constriction when times are tough. So what happens? Well, there's -- first, there's a hiring freeze, then there's some layoffs. And indiscretionary travel, spending goes down and indiscretionary maintenance spending goes down out of all this constriction. And in most companies, when that happens, people hold their breath, they push back, they wine and they just wait for the day when the restrictions are lifted and life can go back as normal. We're not that way at Praxair. There is not this mentality of this bottomless pit of money that we can just spend because we're a highly-profitable enterprise. It's everybody's duty. We jump on cost fast. We over-squeeze to be conservative. There is minimal pushback and exceptions and even less wining about it. My admin snarls at me if I don't book airfare weeks in advance with the lowest carrier. I'm stunned that we even booked this place for this event today, and it's a clear misfire on discipline that we even had sea bass on the menu. That's an expensive entrée. It's just not who we are.
CapEx investing, next topic. Another distinctive part about our culture is our approach CapEx versus expense. So most of what I've seen in my career is that expense dollars get this much scrutiny because that affects the results of today, like thank God that we can capitalize this chunk of spending over here. That's tomorrow's problem, big sign of relief. We're not that way at Praxair because CapEx is precious cash just like expense dollars are precious cash. So we stress test scenarios at what can go wrong in a capital investment and price for the risk and reward in putting in contracts. We avoid the need to. Because when somebody steps forward and says, "This is a strategic investment." That's really a code for the risk return profile stinks, but we want to invest anyway. It's okay to pass on investments. We can grow really well with sub-$2 billion CapEx and a backlog down around 2. We do need to chase everything. It's a smart thing to do. It's the smart thing to be prudent deployers of capital. And here's the real kicker, overruns at Praxair get more scrutiny than anything I've ever seen. A single-digit budget overrun and a few quarters of budget overrun gets an incredible treatment at Praxair. And remember, at a capital-intensive business, on-time and on-budget is everything. And all you have to do is look at the mining industry for the last few years as recent examples of this massive transgressions. I came from a big chemical asset world where overruns would be in the billions and scheduled misses by the years, and I see our organization put more heat and attention in learning around missing this big. And it just sends the right signal to the organization that says, "Every dollar matters. Every week matters. Let's stay on top of this."
Now onto customer negotiations. We have thousands of customer negotiations that take place every year, whether it's for a new asset because we only build a new plant when we have a customer agreement or whether it's an existing contract renewal. And our general approach is we are going to preserve the business model at all expense. So the minimum take-or-pay must be in there. The automatic price escalation must be in there. The contractual protection must be in there, and we're willing to go to court to sue our customer, to protect a single investment, even though the easy path is to let this one go. Because you let one go, you are inviting an erosion of the whole business model. The sale of gas business model is outstanding, but it's outstanding if you stick to it, and you never waver, whether it's in Chicago or in Shanghai. Last, on acquisitions, we generate a lot of cash. We look at a lot of acquisitions. John Panikar shared with you our packaged gas track record, which is outstanding. We buy great a company. We integrate them thoroughly. We're not flawless. sometimes, you think there's an adjacency. You buy a company. You find out it's not really adjacent. We had that experience with the U.S. Homecare business. We missed the operational dynamics and requirements. We missed the pricing dynamics. We sold the business a couple of years ago because it was smarter for us to get out, sell it to somebody who was a natural owner rather than break our pick on dollars and hours to try to fix it. Every company that does acquisitions makes an occasional mistake. Sometimes, it's as simple as overpaying. So the question is do you hide the problem and repeat the mistake, or do you fix the problem and learn from the mistake?
Last set of thoughts in culture, people. So as Eduardo shared earlier on our organizational structure, the single dominant thought on people is that the field P&L trumps headquarters, okay? this is a local business. This is a regional business at best. This is not a global business. And the best way to run it is to push as much empowerment and accountability for the frontline as possible because the product doesn't move. Now a lot of materials company that's based on the nature of assets and the nature of product movement, they have a tendency to globalize, globalize by function, globalize by product line, globalize by asset. But what that tends to invite is an oversized and an overpowered headquarters. Less than 1% of the Praxair employee base works at headquarters, and you're all invited to come visit us. But chances are you won't even be able to find it because there's not a sign on the building. We're a $35 billion market cap company without a sign on the building because the great jobs are on the field, and we like it that way.
Now the point 2 on it, layers. We have layers like any delegation authority and any large enterprise has, but nobody at Praxair mistakes a decentralized business model for a lack of control. We're extremely tight on control, especially on capital spending and on pricing discipline, okay, and on productivity targets. And so it's commonplace for every leader that you're going to hear from today to dive 4 or 5 layers down in the organization and touch the person most in front of a civic situation to find out what's really going on. And that attention to detail and that hands-on style is what drives employee satisfaction because they know that every nickel counts. Now occasionally, if something's out of line, that touch and that employee with the satisfaction turns to dissatisfaction because it's more of a root canal. But that's an occasional necessity.
Onto pay, incentive comp. We have incentive comp plan like a lot of companies. It's a function of overall company, business unit and individual. The formula is most heavily weighted toward profit growth, second is revenue, third is working capital efficiency and we added a return on capital component to encourage capital discipline last year. And maybe that's not anything special or magical. A lot of the companies have that. But what I'm stunned at observing our company is it's formulaic. It's transparent. There's no drama. There's no hidden agenda. The results are undisputed. The formula's known. Here's the bonus payout. You don't like the answer, you push pricing productivity harder next year with your team. There's just not a lot of time and energy wasted on that aspect of winning the business so that the outcome of that is that performers are rewarded, performers are promoted and that's that self-fulfilling prophecy that says, "You want to go up in Praxair, you perform, you get rewarded because you don't get paid for show horse style points." Last point on leadership transition. We have a succession planning process like any big company does. We rack and stack. If they hear with all the people that are ready for this particular leadership job, ready in x years, ready in y years. Antonio and Domy did this interim development opportunity if there's somebody and something else. We do that. A lot of big companies do it.
But let's look at the numbers a little bit. Do we walk the talk? So if you would look at the 17 speakers here today, and you look at their simple profile in your packet, you'll find that 70%, 7-0, are new in the role since 2010 with the company not change organization or strategy. In most large enterprises, if you change 3 quarters of the leaders in a 3-year time period, you would have the strategy zigzagging all over the place. You would have organization changed. Because after all, it's always easier to put leaders in and out when you have organization change as a smokescreen. You would have all type of redirection and communication and efficiency, but not at Praxair. It's quiet. We keep the same strategy. We make the decision, week 1. We announce, week 2. Farewell into week 3, move on. The organization just doesn't spend time on the politics of all that because everybody is just too busy getting their job done. So hopefully, you take away in this slide is that the momentum of the institution far outweighs the impact of any single individual at Praxair. So in closing, I'd put the question back to you and say, does this sound like the kind of culture that can maintain pricing power, that can continue to drive disciplined capital allocation and that can hit productivity targets at twice the rate of our industry peers? Thank you. Now I'd like to introduce Todd Skare, President of Europe.
Todd A. Skare
Thank you, Scott. I've been in Europe since 2010. And you should take out a pencil now because I'm going to tell you something. We've had really not a great economy in Europe since I've been there. But I'm happy to say that Praxair has managed the situation aggressively, really by taking control of what we can control. I can't control the economy, but I can control my cost structure. I can control how we do productivity and where that productivity is directed. And I'm able to stay disciplined in pricing. And those 3 things combined, while the economy in Europe has been really bad in the last few years, Praxair remains very strong in Europe. We've maintained our high operating margins, and we're setup to make extra money if the news that we're hearing about a recovery really takes hold.
Praxair is about $1.4 billion sales company in Europe. That puts us forth in market size compared to our major competitors and a little bit less than 10% market share. I've chosen today to talk about Europe geographically. You heard some of the other businesses talk by business line or different things like that. I think in Europe, it's appropriate to talk geographically because each country is different, has different problems, different opportunities and requires different management priorities to make progress.
So let's look at the map. The first thing you'll see is Praxair is not everywhere in Europe, and that's by design. We believe and you've heard this over and over today that fundamentally, large dense businesses have the best ability to maintain high profit margins and high return on capital. Large businesses can afford the infrastructure and overhead that you need to run a country business. Europe is a bunch of country businesses. You need enough scale to be able to afford those things. Highly-dense businesses have plants nearby each other, customers nearby each other and really opens the door to a lot of productivity, which will drive margins up. So we look very carefully at the geographies that we're in. We decide where to be. We do look at entering new geographies. We don't do it very often, or we make that decision very carefully. And we only do it if we can get a lot of scale and a lot of density to keep the margins up in those businesses.
So the businesses that we do have come in 3 areas. Russia, I'll talk about on the next slide. It's a different -- it's a special case. But the existing businesses are -- about 50% of our sales are in the South, the countries in blue, primarily, Spain and Italy, our largest businesses there. And the green countries of the North have a different situation. We have our businesses in Benelux, Germany and Scandinavia, and I'll talk about those separately. First, in the South. These are Praxair's oldest businesses. They've been in our portfolio the longest amount of time. And it's no secret that those have been the hardest hit business by the economic downturn, bubble of construction in Spain and Italy, government issues persist, and they've really struggled to grow with negative industrial production each of the last 2 years. But these businesses remain strong with operating margins over 20%, and that didn't happen by accident. It was with a lot of hard work. I talked about controlling the things we can control. In Spain and Italy, we decreased our population of employees by 20% in the last 2 years. We have rationalized assets. We had a question about that before lunch. In Spain, we have closed 6 of our 15 cylinder gas filling stations in the last couple of years simply by moving capacity to another place, raising the production levels at the sites that remain and keeping productivity high.
Another one is pricing. One thing is industrial production, and Europe has been negative. But inflation remains persistent and strong. 2%, 3%, 4% inflation across all cost categories in Europe are straight through the recession every year. And you know if you're not able to compensate that inflation with pricing, you know what happens to your margin. We've been able to do pricing. Pricing is mental. You have to deserve the price, and then go ask for the price. And that's simply what we've done. The best example I have is in Italy where we've had 4% positive pricing in Italy, 4%, over the last 3 years simply by going out and going after the price that we deserve.
Next, I'll move to the North, the blue country -- or the green countries. Those countries are newer businesses to Praxair. They came to us mainly through some opportunistic acquisitions that we've made around -- over the last decade, I'll say. You may remember, we bought some pipeline business from one of our competitors in Germany about 10 years ago. That's the big part of our German business today. So that's where that business has come from. They're situated in economies that are better and in areas that we feel we have more opportunities to invest. This is kind of a snapshot of where we are today in Europe x Russia. So how do we go forward from here? In a place like Europe where you hear better news now, I still don't know if I believe it. I hope it's true. But there's a lot of structural problems in Europe, and it would only take a minor thing here or there to slow it back down. So how do we find growth? One is Russia. Another are special areas like the port of Antwerp. We just announced a large project that we're doing along our pipeline network in the port of Antwerp. We're building a large plant there, and that project is all about integration. People wonder, how do you do a project in Belgium during this timeframe? We lined up 2 on-site customers, each taking about half the capacity of the plant, as well as some additional capacity that we needed in our local liquid market, as well as to fill in some of the pipeline customers as we were in a sold-out position. Put that on top with a very nice capital number because we integrated into our existing system, existing pipeline network, and you come up with a very interesting investment case that we took. And I'm very happy to have that project. It's going to change the way we look in the port of Antwerp. Another good example is our investment up North in Scandinavia with Yara. We made a joint venture with Yara called Yara Praxair in 2007 and stepped up in 2011. Before our involvement, Yara was really focused on CO2 and certain businesses that were adjacent to theirs. Now with Praxair's involvement, we're looking at on-site projects to change our mix in that business. We're gaining density in the South of Sweden, and we're looking really hard in oil and gas in the North Sea. We recently had an acquisition of Dominion that you may have read about. There's a dot in Scotland up on the map, probably harder to see. That's the headquarters of Dominion. It's a global company. It's a global offshore company with a lot of activity in the North Sea. That in concert with the network of supply and sales that we have down the Norwegian Coast has us really well-positioned to get into the oil and gas market and grow as that thing grows. Nice thing about that market is it does not go up and down with Europe. It goes up and down globally, and we see a lot of opportunity for us in that area.
So I don't see -- I'm not willing to bet that Europe is going to take off. I'm glad to see that things have flattened out like we talked about this morning, but there are opportunities to invest, and we will invest. We have the money to invest and are willing to invest for the right project that builds density and where we can create extra value. But with the uncertainty, going back to Scott's workhorse, we have a responsibility to continue on the cost side. You'll see us continue to rightsize in the business or product lines that need it, that volumes are low. You'll see us continue to try and consolidate our plans, rationalize assets where we're able to. And then we'll layer on top of that some back-office integration both at a country level and at a region level, which will bring standardization to Europe, which is important and that will find additional productivity. So there are places to grow in Europe. And where we've set ourselves up from a cost position, when the recovery does come and it will eventually come, we're set up to make extra profit.
Next, let's talk about Russia. Russia is a little bit different case, so it's worth a separate discussion. The first question is, why is Praxair in Russia? And I'll give you 4 reasons. Russia has a huge industrial base and a lot of natural resources. But this industrial base is old and needs modernization. It's a government program. It's a program also being taken on by the private sector and is needed. I've been in facilities up in Russia where they have their own air separation plants. And those plants are so old and so low-tech that oftentimes, we can make an investment case for a new facility on energy savings alone. Sprinkle a little bit of growth in there, some integration with the local market or another customer nearby, and you can make really nice investment cases in Russia.
Second, the sale of gas model is being accepted by Russian customers, not their history. Right now, most Russian companies own their own air separation plants. But because they're doing their own modernization, they would prefer to use their capital for that, are accepting a Praxair sale of gas model.
Third, the competitive landscape is different in Russia than we find in other places because there is no local competitor that's strong. So when we go to a new opportunity that we find, a lot of times, we're first in, which is a nice advantage to be able to develop the project the way you want. And the second thing is that even in the final bid, oftentimes, we're only fighting against one other competitor. That's a better situation.
And finally, I'm confident we can build a fully integrated model, as Eduardo talked about this morning. From large on-site that's really our target all the way down to packaged gas, the whole industry -- industrial structure of Russia needs upgrading. Applications that we used to sell in Europe 20 years ago are applicable today in Russia. We just got to go out there and apply those applications, and we'll be able to build a merchant and packaged gas business along with that.
Where are we today? We're operating in the Volga River region right now. That's the circle down towards the southwest side of the map. We have 2 air separation plants that we're running, along with 2 acquisitions that we're busy integrating down in the Volgograd area. Add to that our newly announced joint venture with KuibyshevAzot. That joint venture is going to build a large air separator plant for a new ammonia production. And then up in the Urals region, which will start up next year, we'll be starting up our anchor plant at NTMK, which is a large steel mill in EVRAZ near Ekaterinburg. That will start up next year and get that region for Russia started.
Now what's going to happen in the future in Russia? This is what we're doing now. The good news is it is opportunity-rich. I'm looking right now at a list of opportunities that will allow us to keep this backlog for years to come, probably for the next decade. The nice thing about a long list is that we can be careful. We can pick the areas we want to be in, the customers we want to be with, and we can find the right risk and reward balance for the projects that we're looking for.
I talked about risk. It's true. It's probably going through some of your minds. Russia is a risky place to be, and that's true. It's a difficult place to be, and the highest-risk part that we found in Russia is project execution. If you interview other companies that are working in Russia, doing start-ups in Russia, it's hard to do projects.
That's one of the -- I'm an old project guy, and it's one of the hardest places I've seen to do projects. What we've done to improve that -- and I see the improvement coming already, as we've decided to get local with our execution, something we've done in Shanghai in China, and it's worked fantastic -- the simple premise that local people know how to do things locally.
So we're building a team in Moscow with Russia-specific skills, like permitting, construction contracting, construction management, transportation. This team is now executing their third project, and you can see the dramatic improvements that we're making. And I'm confident that we're building a competitive advantage in project execution that's going to allow us to take more than our fair share of these opportunities that I talked about.
So I'm excited about Russia. Compared to when I went up there, the number of opportunities are a lot more than what I thought. It's true that execution is tough, but we're solving that problem. And Praxair's track record for solving tough execution problems, I'm confident, is going to give us an edge going forward.
Thank you. Next, I'll invite Domingos Bulus up. He's the President of our valuable company in South America.
Domingos Henrique Guimarães Bulus
Good afternoon, ladies and gentlemen. Just it's a pleasure to be here today and have the opportunity to talk about South America. South America is a region where Praxair has its second largest operation and also where it's one of its core geographies.
I've been engaged with this corporate for 29 years. And I'm back to South America, specifically Brazil in the last 10 years. So I'm a little bit used with the geographies of very complex geography. But sometimes, geography, it's a complex environment.
So just to give you the -- our size, there is a market that -- about $4.5 billion industrial gas markets, where Praxair has around 50%. We are #1 or #2 in most countries, except in Chile where we are #3. 8% of the -- of our revenue come out of Brazil, and the balance is basically concentrated in 5 top countries, such as Argentina, Colombia, Peru, Chile and Venezuela.
Along the years, along these 100 years, in Brazil, we have built a very solid infrastructure and over 20 years in the South American countries. We have in place nowadays about 5,500 members. I would say a kind of work for us that is very eager to understand and go after every opportunity internally in terms of productivity, available and also in terms of the market field.
Likewise, Eduardo has mentioned the morning in terms of the North America organization. We have a fully integrated business model in Brazil and also in South America. We are replicating everything that we have in U.S. and in Brazil to other countries of America. So we get our site package in bulk down there. We get density. We get network, very strong network, reliability and focus. And more than that, we have attitude. It means that which are the key elements of our strategy to serve the industry in the geography.
Each of the territory is closely monitored as we defend the castle tow. It's basically we -- there is no room uncovered in South America that we need all this own niche. We play niche. We play end user all the time.
Just to give you an idea about our dimension. We deliver -- we make about 80,000-plus product deliveries a day and have about 10,000 customer interface. Just to give you an idea that I mentioned of our operation and the opportunity that we have to understand the markets, to understand a better -- and anticipate wretch [ph], risks and opportunities.
Our local application thing and engineering manufacturing capability just is a kind of complement our technical approach to serve the industry, to serve our customers down there. Basically, these are the reasons that we have -- that we enjoy a 99% retention rate, order-induced, and we have about over 70% rerate [ph] in all those site visits.
As you know, South America is a region of challenge and diversities and all -- but also great opportunities. Just let's talk about the challenge that we have. We've been having, for a long time, currency volatility, high inflation, high costs to serve. I talk about not only power costs but also transportation costs, poor infrastructure, and bottlenecks are not new -- none of them are new for us. We know how to handle that. We are used to -- and usually, we consider that in our business case.
Let's talk about some opportunities. In general, South America, from Colombia up north to Argentina down the south, is a region of vast mineral and hydrocarbon resource. The big question is when we're going to have all of the resource capitalized and monetized and see baked in our case, when we'll have these available to sell more gases? It's a big question. So we -- what we face there is slow movement, slow pace for the government decisions. All the products have been postponed, and we don't get many activity nowadays.
But we do have some private opportunities related, such as -- let's give an example. Let's take energy, for example. Biofuels. We have many plants down in Argentina, Colombia and Brazil being done and have opportunities to sell nitrogen, for nitrozation for bracketing or pH control, many opportunity that we have. Let's take, for example, power plants. There's several new power plants spread out mainly in Brazil that we have opportunity to sell bulk in packaged gas.
Let's take shipyard industry. They probably has put a plant in place to increase the local content for the industry. This good. It's positive. We have 10 to 15 new shipyards foreign investment plus a bunch of small and local investment in Brazil that we are selling a bunch of gas. So we don't have this in the -- rather squint [ph] as a big opportunity to leverage our sale but give us many fees that will give us end of the day very good results.
Let's talk about, for example, we're going to have own bid in the next couple of months. Some 1 or 2 refineries, as mentioned the morning, in Latin -- in South America to bid and have an opportunity to also to win that infrastructure.
We -- a lot of conversation we -- related to the World Cup, to the Olympic Games, but we have not considered this in our plan at this point. Many products are, I would say, late, are delayed, And we believe that the next 2 years, we're going to get some benefit, and we are consider this as an upside. We can't consider this in the base case. Otherwise, we will not delivered our results. For -- take for example, on the consumption side, it will take just the middle class. Middle-class don't [indiscernible] outside of Brazil, 100 million people and also inside of Brazil, under 40 million people that are increasing the high style of living, and they have much more purchase power nowadays than they used to be.
So there are lots of pressure on the consumption, also the local industry. This important for us. We see several opportunities in the manufacturing industry. And I give you example. For example, there are 8 new other plants, foreign investment in Brazil under construction right now and have more bulk and packaged sales for us. It's not only Asia but European, also U.S. companies. They are moving there to the side, all of market in Brazil, that sizable market, about 3 million to 3.5 million vehicles a year.
We also see opportunities in terms of these 2 producer. They are -- they have enough capacity there. They need to recover their export sales, and they are introducing new special products. So we give responsibility to sell more nitrogen or argon, and also, they need more productivity to be more competitive. So it's something that we have technology -- we have gas intense technology available to offer to them.
We -- if you take a look in this graph here, that shows you our history in the last 10 years. We were able to grow about, in local currency, 9% against IP, average IP at 3% per year. So it's a interesting growth, and we are consider not only to repeat that but even exceed this in the next coming years.
How are we going to achieve that? It's one point that's important is application technology. We plan -- we -- just imagine with me, for example, all the market is very open in Brazil. The local industry open to receive new technology, to receive all the best practice that the industry has -- are all developed. What we do is just the same way that we have, weld mixtures being sold in U.S., laser-cutting practice that to be implemented in U.S. We use in Brazil. We use in Peru, in Chile or in Colombia just to -- as a kind of implement new application, new technology and is very well perceived by the customers.
The second point is productivity. As been said here, in terms of productivity, it's not only relevant for us to deliver, but it's mandatory. And we work -- at least we consider global perspective is up 5% cost stack each year that we need to reduce, but we need much more than that. So it's relevant to move away for the comfort zone in every activity that we perform in South America.
Another point, to recover price -- fully recover cost inflation through price. This one thing that we can't put this aside is part of the strategy. And every contract detail, supply need, everything we need to allow to get more price, to get more contract, better contract negotiation, better leverage with the contract negotiation. What we do is we hit the scorecard. We need KPIs where we can measure and understand exactly what kind of customer industry that we are, negotiating to take and get the best or the major benefit out of the contracts.
So in general, what we have is a challenged situation going forward, and what we are considering that if we are able to perform all the activities that we've been doing the last years, also keeping and growing the market share as we move, not only Brazil, outside Brazil, all the Hispanic countries, sure, we're going to deliver the results that the corporation expect from us.
With this, that's all I have today. I turn to Dr. Anne Roby, Praxair Asia President.
Anne K. Roby
Thank you, Domingos. We've talked a lot today about density and discipline. And I'd like to talk to you this afternoon about how that applies particularly in Asia because I think it's a great example of where we need to show this commitment to these values.
So I'll focus most of my talk on China. As you see in the upper left-hand corner, China, although it, over the past 20 years, has gone through really an industrial revolution with industrial production growing by a factor of 10, it still is only about 20% on a per capita basis in the consumption of industrial gases. So we still have many, many years, many decades of growth left in China.
Now looking at where we are in China in the macro environment today, clearly, we are seeing slower growth. We're moving from high double-digit growth over this time horizon to high single-digit growth. And it's really going to be a combination of 2 dynamics. The new administration in China is laser-focused on rebalancing the economy. Basically, China is paying for the last 20 years of explosive growth with issues like environmental impact and it -- distribution of its people, right, in its distribution of its people. So it's focused very much on rebalancing, in particular, the industrial sector.
What we'll see over the next 5 to 10 years is the shutting down of some of the most old -- some of the oldest and most polluting assets. But we'll still see continued new projects. The government policy is driving environmental compliance, which feeds very well into Praxair's core capabilities and also driving more technology-intensive industries. We also are seeing faster growth in land as the government tries to rebalance its population.
So as Dan mentioned before, we have a lot of opportunities in refining industry in China as the new fuel standards come into play. That drives hydrogen demand, and as Dan mentioned, in China, that will be delivered through coal gasification.
We also see continued growth in chemicals market, again, primarily through coal gasification. What we've seen in the last 2 years, in particular amongst our customers in China, is a real pull for some of our applications technology, particularly when we can bring energy efficiency and when we can bring environmental solutions. I think in China, we have a higher density of our new opportunities that are coming from environmental applications than anywhere else around the world.
So as we put all of this together and we look at the horizon for the next 5 years, about half of our growth will be coming from loading up our existing base assets, including our merchant liquids and half from new projects, primarily coal gasification, leading to a compound growth rate of about 15%. Not all of our new projects have been announced, and we'll be talking about some of them.
So as I mentioned, we've talked a lot this morning about our density strategy. I think over the last 20 years, China gives us the best example of how density strategy correlates with operating margin. So Praxair China has 50 plants or projects that are currently underway. And as you can see, there's a high density in 3 locations along the coast around Beijing, between Shanghai and Nanjing, and then in the south in Guangzhou province. You see one outlier there. That's in Chongqing, and you can take it from our existing strategy, that probably won't be our last plant in Chongqing.
The green bar there represents our relative operating profit margin. And Praxair is the leading by this -- leading industrial gas supplier by this metric in China. You look at our next closest competitor. You can see that they're primarily coastal but certainly more dispersed than Praxair. You can also see that their operating profit margin is a little bit less. You look at our next, even further dispersion of their assets. And finally, our lagging competitor will pretty much invest everywhere across country with no density at all. So it's clear that the correlation is direct, and I think it's the best example we have anywhere in the world of how correlated these 2 elements are.
So we also talked a lot this morning around discipline and the discipline around project selection. Right now in China, you've heard a lot about coal gasification, driving several different industries. Dan mentioned the hydrogen industry. Coal gasification, simply put, is taking coal from a mine, processing it, reacting it with oxygen and making syngas or CO and hydrogen. The same syngas that you can make from a syn-methane reformer, and it can yield the same product slate of refining hydrogen, fuel, liquid fuels, methanol and downstream chemicals or ammonia. So same feedstock -- or different feedstock, same seed slate.
Right now in China, there is more than 200 projects being developed in coal gasification that are expected to start up between now and 2020, 200 projects that from an industrial gas perspective, would yield about $20 billion of investment. Now not all of these projects are going to happen. So as we look at where we spend our efforts in trying to gain new business, the first things we look at are, is the project really viable? Is the end market, the ammonia or the methanol or the hydrogen needed where this project is being developed? Does the operator have any experience? Do they know what they're doing or is this the first time they've ever built or operated a coal gasifier? Do they have access to water? What is the likelihood that this project will be permitted?
So we really need to become experts, not just in oxygen plants or oxygen production but also in coal gasification in the markets that it serves. Then we filter it through our normal Praxair considerations. Is this a good customer? Is this a customer that's going to pay his bills, not just today but for the next 20 years? Does this fit our density model. Is this an outlier in a province that we've never been and will probably never go again? And is the scope of the project and the oxygen demand and the product line a good fit for what we can do?
So as we look at these 200-some projects, we still trim them down and bid on really less than 10% of them. And as you can see, there's been many projects announced recently. Praxair has been awarded several projects we haven't yet announced but will shortly once we finalize all the terms and conditions.
We do have 2 plants that are operating right now and have been operating for several years. SOPO outside of Shanghai, which produces acetic acid, is the first large oxygen supply over the fence to a coal gasification process. Wuwei in Anhui province is the second. There haven't been any other start-ups, though Praxair is the only industrial gas supplier supplying oxygen over the fence to a coal gasifier.
We have another project in development in Yangquan that will start up in 2015. So we have excellent operational experience. And in fact, that operational experience has caused a lot of producers to come to us as their preferred supplier because we have that operational experience.
Talk a bit more about outside of China and what we're seeing. Our oldest business in Asia is in Thailand. We've been in Thailand for 40 years, another great example of a density strategy where last year, we started up our 6th CO2 plant in Rayong.
We also have a significant business in Korea. We announced recently a new plant starting up to supply Samsung, and we have 2 other projects under construction now to, again, supply Samsung's growth. And our success in our business in Korea really has been selecting the right company to supply. We are the largest industrial gas supplier to Samsung, and our growth has really benefited from their growth.
Finally, I'll talk about India. Now India, much like Brazil, is a very complicated country in which to do business and certainly has some headwinds currently with the devaluation. And I would've said a year ago that we wouldn't have many more opportunities in India long term, but we're actually finding we have a very healthy pipeline of opportunities that are not driven by new plant start-ups, new project start-ups but really driven by our ability to help us -- help make our customers more competitive by new application technology, particularly for the steel industry, that is really driving a lot of our growth in the future for India. And Ray will talk a little bit more about that.
But as you can see throughout Asia, Asia for Praxair is not all of Asia. It's really 4 countries where we focus our investment, and by doing that, we've been able to grow profitably.
So I will turn it over to Ray to talk about some of our applications, technology and our new developments.
So the next 3 speakers, myself, I'll be talking about technology. And then Ben Glazer will talk about capital projects and how we approve capital projects, how we allocate capital. And Murray will talk about project execution. When we win a new plant contract, how do we execute that project? And all 3 of these functions are run as global functions within Praxair to really take advantage of a common technology platform, a common decision process around what projects to go after, what our return criteria are and a common process for executing those. But like everything in Praxair, we really work very, very closely with the regions, and most of the work gets done locally. But whether it's applications technology or project execution, the work really happens locally.
First, let me talk about the 3 key things we work on in technology. It's really pretty straightforward. We focus on supply systems and continuing to improve our supply technology. We focus on productivity, and you've heard a lot about productivity today. And thirdly, growth through applications technology.
In supply systems, there are technology programs, both for air separation systems as well as for hydrogen units. And in both cases, our goal and our achievement is to reduce the cost of products supplied by supply systems by 3% every year. So we adopted this goal about 5 years ago. We put in a very detailed project plan to develop technology, to insert that technology into our plants and to achieve at least 3% lower cost of products every year through new technology deployment.
We've also focused a lot on what we call product line platforms in our supply systems. And my next slide goes into that in more detail. But you can think about this like Boeing, for example, develops the 737 platform. And they do all of the engineering work associated with that platform once. And they build some flexibility into that fleet, and then they get to use that platform over and over again, making, in their case, hundreds or thousands of planes with the same platform. So we have the same concept in Praxair, and again, our goal here is to be able to reuse the engineering, get very good at that platform and achieve our 3% lower cost every year.
Productivity. We talked a lot about productivity all day. I think you recognize it's part of our culture. Sean Durbin talked about it this morning, and he gave you several examples of specific productivity programs that we have deployed. In technology, what we do in productivity is really to develop new trouble machinery technology, new control technology, using operational research tools to basically take all of the data we're getting from all the customers we supply, all the data from our onboard computers to figure out how to improve our distribution efficiency.
And frankly, every year, we're inserting new technology into the system. The way we organize productivity is we have, first of all, at a region and a company basis, all 28,000 employs are focused on productivity. All businesses have goals around productivity, but we do have hundreds of specialists whose full-time job is to develop new productivity offerings, better operational tools, better optimization tools, better control systems and to work with the organization to deploy those tools globally.
Third, applications technologies. Here, it's about growth. So our whole purpose here, our whole objective here is to continue to work with the market segments we serve to understand their needs, to improve their productivity as customers, the quality of their products, the energy efficiency of their operations, the environmental discharge from their operations and to work with market segments and customers to bring them value-added technology to help them reduce their costs, reduce their environmental footprint, improve their product quality and so forth.
So we've got hundreds of people, both in a central R&D organization but also spread throughout the world, who work with customers and applications technology and bringing new solutions, which drives our growth. So again, just to frame this, in supply systems, about 1,200, 1,500 people working globally, enduring professionals in R&D to improve supply systems, hundreds of people specializing in productivity, working with 27,000 people to deliver productivity results, and then applications technology of both R&D as well as applications specialists in every country to deliver top line growth.
Let me say some more about our product line approach in supply systems. We started this, I think, about 8 years ago, 8 to 10 years ago, and we recognize that we were spending too much time custom-designing plants as specific opportunities came up, and that it was very risky because when you do a design, a unique design, a new design for a project, it's time-consuming for the engineering, and then you're trying new things and some of those may work, some of them may not. So we adopted a product line strategy in which we work with our businesses, work with the market segments to understand what are some common needs, what are the most likely to be used plant size is, what kind of product mix do those plants have to produce? And so how much oxygen, how much nitrogen, how much argon? How much flexibility do we have to have in terms of demand of those products? Pressure, purity, those sorts of issues.
So you look at a set of opportunities in a side range and we design a product line platform that has a firm amount of flexibility in terms of being able to meet a group of needs. We do the engineering around that platform. We insert new technology into that platform to make sure it's very efficient, and then we replicate that platform at every opportunity we can that fits within that envelope. So the advantages here are we get rapid learning, we learn what's working well and if there are any problems in that first plant, we correct it in the second plant, and then we execute multiple projects with the same plant design. Engineering is done once and then the rest of the engineering on a project basis is limited to the specifics of that project. We get a lot of procurement leverage. We're buying basically the same components for that product line plant over and over again, so suppliers give us better pricing.
And lastly from an availability, reliability viewpoint, we're able to afford to have spare parts. So all the major components that go into a product line plant, we have global spares for that, in some case, regional spares. So if a compressor fails, if a turbine files, if a valve fails, we have the ability to spare those parts and to be able to get that plant up and running very, very quickly at very low cost for the organization.
I'm showing 3 bar charts on the bottom of this page to give you a sense of our progress on product line plants and the opportunity we continue to have to improve here. In 2008, we had about 10 distinct product lines, platforms that we used. By 2013, this year, we have 15 -- so we've increased the density, we've increased the number of product line platforms that we have. 2018, we will have 20. So we've identified where we have holes, where there are some new market needs that we want to serve better, and we are creating 5 new product line platforms to meet those new needs.
If you look at the number of plants that were in operation in 2008 that used product lines, we had 35 in 2008. Today, we have 90. So we're using these product line plants much more than we were in the past. Much more in the past was custom, now it's product line platforms. And then as a percentage of our total CapEx, you can see that 2008, it was 60%. By 2018, it will be more than 90%, of our capital expenditures will use product line platforms. And for us, that means much lower risk, better capital efficiency.
We think it's a real competitive advantage that we have in terms of how we're designing plants and executing those projects. Murray will talk some more about how we execute these projects in a couple of presentations.
Let me switch gears to applications technology, and many of the speakers responsible for the regional businesses have talked about the importance of this. But what we're doing here again is to develop solutions and to make those solutions available to customers that use industrial gases to help a customer achieve their goals. The chart on the right is the point that Anne made, and it's that if you compare the developing world with mature markets like the United States, even a country like China only uses about 20% of the industrial gases that we use in the United States on a per-unit operations basis. So if you look at, for example, how much oxygen is used per ton of steel produced, or how much argon is used per pound of welding done, or you can go on and on in terms of the intensity of industrial gas use by country. There's a lot of opportunity for growth in the emerging economies. So when we talk about growth, we talk about it mainly in the 3 Es, emerging economies, the energy sector and environmental. And frankly, all 3 of those sectors are -- those 3 of those drivers are working well for us in emerging economies, but they're also working well in the developed economies.
We've highlighted on the left side of this page, several key drivers. China has a huge opportunity in terms of improving their environmental footprint. They're putting in programs around air quality. Those really benefit us. As you switch from doing something with air, making steel with an airbase process versus oxygen, combustion processes, glass -- making glass with oxy-fuel burners instead of air, significant improvement in both CO2 emissions as well as NOx emissions. And so we have a lot of drivers around environmental, and I can go down the list.
So we see in all the geographies that we are working in, significant opportunity in -- through applications technology. My last chart gives you a specific example on steel, and we're showing here the amount of oxygen used to make steel, the amount of -- the intensity of oxygen used in steelmaking from 1995 through the present. And you can see that today, steel companies are using about 50% more oxygen per ton of steel than they were using in '95. On the bottom in blue, we've highlighted several technologies where we proactively came to the market, developed an application, brought that application, brought that technology to the marketplace. And on the top of the curve, it's some of the things that drove this change. In the '90s, it was steel companies looking to improve the productivity of their blast furnaces and by using more oxygen, they were able to increase the throughput for blast furnace.
Then they wanted to go to lower-cost fuels. So they wanted to go from coke to powdered coal and now to natural gas. And each one of those substitutions required more industrial gases. So when we think about applications technology, when we think about technology in general, from supply systems, to productivity, to applications, it's really about the theme that Praxair has around making our planet more productive, and I think we're delivering those results to customers as well as internally.
So with that, I'll turn it over to Ben Glazer, who will be talking about our process for capital decision-making.
Good afternoon. My name is Ben Glazer, and part of my responsibilities in supporting our senior leadership team include how we manage our capital projects and select our capital projects. And as most of you are aware, we exist and thrive in an industry that's very capital-intensive. To maintain our operations, to grow our business, it requires a significant amount of investment, and as you heard Steve talk about earlier, we spend a considerable amount of time around these decisions. For our part, we spend approximately $2 billion a year in capital investments and we spend that time to make sure that we get it right.
Now many of the things I'm going to share with you today, they're not necessarily uncommon, they're not even unique, but really, what differentiates ourselves or differentiates us, what makes us special is our rigorous and consistent application of these programs and processes and ideas. It's our discipline. It's what makes us different and it's what ultimately allows us to be successful and achieve the numbers that you've seen here today. So with that, let me get right into it.
And the first slide that I'm showing here really is a basic outline of our capital selection process. It's a very simple gated-review process where each one of these state's gates has an entry, an exit criteria associated with it. To start with, as we consider the projects that we identify, the business that we like to develop, the opportunities that we go after, we like to make sure that it aligns with the strategies that we've outlined as a corporation: building out integrated supply systems of on-site merchant and package; leveraging a sale of gas business model in our targeted geographies; and building out density of operations where we have existing assets and expertise on the ground that we can leverage. Okay?
Now as we continue to move to this process, we have things like heads-up reviews. We use those for initial screenings and assessments of projects, to make sure that we're not going to waste time on projects or waste resource on projects that have low profitability, that it's just evident from the onset that there's substantial risk associated with that project and not returns in exchange for it. And then as we continue through that approval process, one of the things that we do is really employ a consistent set of standards for every project along the way, no exceptions. Okay? To start with, we use a standard global hurdle rate, a hurdle rate applied to all the business and regions that we operate in around the world. We generally set that at about 500 basis points above our cost of capital, and then helps us align with our overall ROC goals as a corporation. We also use a standard discounted cash flow model that centrally control and manage everything, from the assumptions that go into it, the modeling techniques, how it's implemented and reviewed, all centrally controlled by my group. So for example, things like foreign exchange rates, devaluation, inflation, taxes and so on, is all controlled centrally to make sure that we have consistent implementation of these assumptions for any project, anywhere around the world that we operate in.
As you might expect, we have thorough legal terms of conditions review with any project to make sure we don't have any undue liability or undue risk associated with the project. And then finally, every project goes through a thorough risk-based return evaluation. I'm going to spend some more time on that because I think it's worth talking about. And finally, once we have a project approved, this process is not complete. As project moves from design to execution, to startup, there's regular manager of use associated with that. You're going to hear Murray Covello get up here and talk about how that engineering and execution process really worked hand-in-hand with our procedures and overall philosophy of how we select projects.
And then finally, we have various audits for any of the investments that we make, whether that's looking at operational performance of the plant, compared to its original design criteria, or certainly, the economic performance of a plant compared to the original economic assumptions that the project was approved in -- originally approved upon. And all that information circle back -- sent then back to the various organizations and businesses, so we can make continuous learning from it.
So let me now move and share a little bit more detail in what I mean by this. Return-based investment criteria. So we recognize that not all projects are created equal. Each project is an individual with individual sets of risks, opportunities and attributes. And we spend considerable time talking about those risks and attributes and the associated returns and trade-offs, and this is 1 tool that really helps us do that and frame that. So what I'm showing here on the left-hand side in blue is a set of criteria or investment conditions that we look at. And we're able to grade those criteria or investment considerations across a spectrum of risk. Depending on where it falls in that spectrum, we can make certain adjustments to our hurdle rate, to account for the risk and opportunities associated with the project.
So let me talk through an example. One of the first items you see there is guaranteed cash flows. What I mean by that is what is the certainty of the future cash flows associated with the project? Okay? What is it that we can take to a bank with that project? So you see the first 2 items are cost-reduction, take-or-pay. Those are things that we consider have a fairly high degree of certainty. For cost-reduction projects, we're investing capital to go ahead and get reduced our operating costs or lower our energy consumption. And whether it's a case of the capital expenditures or the paybacks associated with that, we're generally in control. And so there's a fair degree of certainty of that payback. In take-or-pay, you heard earlier, these are where customers are guaranteeing a minimum monthly revenue, regardless of how much volume they take. Okay? So again, a fair degree of certainty there.
As you move down the spectrum, there are project investments where these plants make certain co-product liquids that are then sold into the merchant market. Those cash flows are dependent upon getting new customers in the merchant market that are not usually or not -- or may not be entirely secured when the original project investment is made. Okay, so there's more variability associated with merchant credits.
And so we, like at any given project, look at the mix of contributions from each one of these categories and we can begin to make decisions from that. We move on down the list, we consider customer counter-party risk. We're just not talking about the financial strength of the customer. You heard Anne mention, we spend a lot of time going in and understanding our customers' operations and assets whether that's in steel, petrochemicals and refinery, how competitive are they in their respective markets? We consider sovereign risk, what's the foreign exchange situation like, what's been our track record, or operating in a given country, our ability to enforce our contracts, and then certainly, execution risk. Is it the first time that we built or designed a plant? Or is it a product line plant that you heard Ray spoke about that we've built, owned and operated a dozen times the world over, and then certainly, in alignment with our strategy.
So ultimately, what I hope you see here is a set of procedures and processes that we really employ a great deal of rigor and consistent application and focus. That's very important and instrumental to help our senior leadership ultimately make these decisions. But what I find more important is that it sets a tone. It sets a tone, it sets a culture around our organization. What I can tell you, every single Praxair employee in this room, every person that owns a P&L, every business development manager, every controller, every purchasing agent and so far down the line, understand how projects are evaluated and they know, as a corporation, what we value. And what that means is that we have projects that are high-quality and that we can deliver with a high degree of certainty.
So how does this play out? This next slide, I'll share an example of what's a representative investment. And with that, I mean, investment you have, a period of time that we're actually making that investment, building out capital, putting a construction progress on the balance sheet, and as you might expect during that time, you're probably going to have relatively flat return on capital to maybe, perhaps, slightly negative return on capital. And I should have mentioned this is a return on capital slide. And as you move into the, really, the initial contract term, there's a couple of things that are going on. Okay? The plant starts up, so you're beginning to earn revenues, facility fees from those take-or-pays, and what makes up the numerator of return of capital, it's NOPAT or net operating profit after taxes.
And NOPAT, you have initial startup, and begins to really incrementally increase over time. That plant is being loaded, but you also have these facility fees and take-or-pay contracts that you heard Scott Kaltrider speak about earlier that are being escalated, escalated by inflation and other formulas that continue to incrementally increase even following plant startup. Now on the denominator side, the capital side, this is our investment, things begin to move from construction in progress over to our fixed asset register, and they begin to depreciate. They depreciate over this initial economic term, typically 15-years cycle. And so you have this incrementally increasing numerator, this declining balance of the denominator, and what you end up with is really an exceedingly increasing return on capital throughout this initial term. We generally think that we are accretive to our corporate return on capital within 5 years, and by the end of the term, you have these return on capital numbers that are very, very high.
And then we have the opportunity to actually renew that investment, and we have a very good track record of renewing that investment. It's not uncommon for on-site contracts, for these original investments to go through multiple successive renewal terms. Okay? And while we may need to make periodic investments in plant equipment, okay, to go ahead and go through these renewal terms, you're still dealing with a planned investment with a largely depreciated capital base, sunken leverage engineering costs as a result of our product line strategy, and that's incredibly important that we take seriously. And we were able to leverage that. I think that's evident in our return on capital numbers and it's evident in our operating profit performance.
The last slide that I'm going to leave you with is really an outlook and a representation of backlog by our industry. And I've normalized this to roughly 100% and scaled it associated with geography. And Praxair's in the first column and then I'm showing Praxair as I expect it to be at the end of 2013. I've taken out some of the larger investments associated with the hydrogen projects. And I think what you see here is a fairly balanced and diversed set of backlog project opportunities right across our targeted geographies, where we've arrived at that through careful selection in these processes that I've explained to you.
If you then move forward and look at our competitors, I think one of the things that jumps out to you is a real investment concentration, particularly in China. Okay, this investment concentration is really if you look at the details, it's really a product of really a few large project investments related to coal gasification. Now you heard Anne Roby get up here and speak earlier about the opportunities associated with coal gasification, but also she spoke about the careful, measured and careful considerations that we go through in looking at and evaluating these projects. These projects are not all created equal.
As you go ahead and think about what this means for Praxair, Praxair is going to continue to be in the coal gasification space. We've made investments in these, and we're going to continue to make investments. But it's going to be with very careful and concerned selective programs and processes that I've outlined before. I think what you can conclude from this slide is a couple of things. One, I think it's arguably that Praxair has the most balanced and diversed backlog across the industry; anchored, in a large part, in the Americas where we have very good sound assets and operations strength and a continued positive outlook; Europe is a large representation of the investments you heard Todd talk about in Russia as well as bolting on to some of our footholds in Northern Europe; and then finally in Asia, where we have really a wealth of opportunities that's really a product of a very careful selection process that you heard Anne speak about earlier.
So ultimately, you as an analyst will make your own conclusion about what this means for Praxair and our competitors, but for our part, you can be sure that this process of rigorous discipline, enforcement of these processes and procedures I spoke about, that we'll continue to adhere to. And I believe that it's one of our secrets to success and ultimately, will play out in our numbers long-term.
And so with that, I'm going to introduce Murray Covello. He's our Vice President of Global Supply Systems, and has a considerable job of really engineering and executing this project backlog.
Murray G. Covello
Thank you, Ben. Well, it's my pleasure this afternoon to talk to you about the exciting world of project execution. The reason it's exciting is from lots of the presentations you've seen, I hope you recognize that we get to travel to very interesting places around the world on a regular occasion, and the other thing that makes it exciting is when it goes poorly. Scott Telesz mentioned if you don't perform in a capital-execution environment in a capital-intensive business like Praxair, things could go bad very, very quickly.
And so to get good at this is taking a fair amount of time for us, and we've adopted some excellent processes and discipline to get where we need to be. When you look at $2 billion of portfolio, that represents over 100 projects around the world that have to be done day-in, day-out with excellence, and with people in many different countries operating in tough environments and with tough challenges everywhere they go to execute these projects. So hopefully over the next few slides, I'll be able to give you some sense of the processes we use, the deployment of our team and then finally, what I consider world-class results and something that's very difficult to do and we're very proud of, as we look at continued excellence in the capital-project environment.
So looking at the front-end loading process. This is the FEL process, adopted from a company called Independent Project Analysis, really a company that looks at the capital environment, and has developed this process that we adopted around 2000 and have spent a number of years looking at applying across the portfolio of projects. We do not do any project in Praxair that doesn't go through the FEL-2 to solidify the expectations that we have within the project and for the business. And you can see the time sequence here is obviously going through the budgetary stages with the customer and with our business, to understand what's being asked for a given supply situation. And what can we do around the concepts of building on the product line definitions that we have, but also adding in all of the infrastructure, or all of the options we might want to add in terms of argon production or pipelines or anything else that might be required to make this system work.
And then moving into FEL-1 is where we cycle a number of alternatives and get into some initial estimates, look at some design iterations in terms of concepts and which product lines might fit best into a given situation, and really start to look at optimizing the supply solutions for a different -- a given alternative or given situation. As much as we talk about the product line as forming the core of the plants or the facility that's required, there's lots of engineering decisions and lots of things that have to be decided in terms of all of the utilities and other things that go into a project to make a complete package in terms of bringing that to business leadership.
Moving into FEL-2 is where we spend the bulk of our time. Within the planning process, probably 80% of our time is spent on FEL-2, really getting into the details. And I liken this to building a house. Right? If you'd go into the planning cycle for building a house and hand that off to a contractor without a lot of detail, it will be very exciting and very costly. So within the FEL-2 process, we're really looking to bring a controlled cost estimate for a single solution, a very good detail, and working hand-in-hand with business to finalize all of the decisions on scope and specifications, to look at our risk plans for a given geography and make sure we understand what might go wrong, but at the same time, also look at opportunities where things can be changed or over the life of the project, we may get some opportunities to improve the performance of the project.
We work hand-in-hand with our procurement team and during FEL-2, we expect more than 80% of our equipment to be quoted. We expect there to be labor surveys and information obtained for a given location where we're going to go and execute a project, so we have a high degree of certainty; what our exposure is to productivity, labor rates, construction availability and equipment we're going to use for a project.
And then finally, this is not just for the capital solution. We are integrated with our operations team for them to assist in making decisions that are going to guarantee the lowest lifecycle cost for that 15 or 20 or 25 years, when that plant's going to be unearthed. That's the point after FEL-2 that we're going to get this project approved and sanctioned by management, and we're going to make our commitments to the customer and then moving into project execution.
We move into something towards FEL-3 and it's another stage-gate, where we've made all our key purchases, and then we're able to approve that stage-gate and move into detailed design and really finish off the construction of the project -- or sorry, the design of the project to pass construction packages off to contractors for the facility and get it built. And then move into startup with our operations team, and it doesn't just stop there. Because we have an obligation to the first year reliability of that plant before it really transfers operation, and we have a commitment to hit our design target for power and production capacity within 1% of design. So we are committed to meet the expectations of the operating group. And then finally, as Ben mentioned, we have an audit process after the plant's been running for a year period to really go back and look at the complete package of the capital cost, energy efficiency, production capability of the plant to make sure it met business expectations. So really, when you look at the careful selection process that Ben talked about, it flows into this process we have. We're really managing tightly the concept of the planning cycle and the execution cycle for any capital project in Praxair. When you look at doing a project, it really ends up being about people. I think you heard Steve talk about that this morning. That in the end, for any group, it's about people. I have over 1,000 people around the world in the GSS organization. These people are out there managing the design, engineering, construction project management, all the things that go hand-in-hand into building the facilities for Praxair. We're not looking at an SOE environment where we're generating equipment for sale. We're not doing this as a profit center. This is solely to be aligned with the business organization in Praxair. You can see the layout of the people, how they're deployed around the world, fully aligned with the regions you heard speak today and they're fully aligned, not only because they're located in those regions, but they're administratively paid and compensated in the same way as the people are in the business organizations in that country or region and that gives us full alignment and full connectivity with those business groups. When you look at the breakdown, you can see that we have a large amount of our design activity in the U.S., very concerned about managing intellectual property, but we're also looking at the advantages we have to be co-located with development teams that Rey spoke about, who are located in our office, so our engineering teams and our research and development teams for supply systems and the operating groups for global operations are co-located, that gives us some strong synergies to end up with great designs that are coming out of this U.S. team. After that, it really is about execution organization. You heard Todd speak about the strength of having local expertise. So people understand the local language, the local standards and practices around contracting and how to get things done in a given country. And when we go down to our second biggest organization in China, not only are they geared towards executing for China, but they also are a key region to supply engineering capability for the other countries in Asia, but also for Russia and other geographies where they have the ability to generate non-IP sensitive engineering that gives us a value center for engineering capability. We're very happy with the expansion of our Russian team as Todd talked about, not only his business organization, but the engineering team has grown very well. Very strong in terms of local talent. We're very pleased with the capabilities they're developing and you can see we're on our fourth project in Russia, fourth significant project in Russia. And we're looking forward to the capabilities that team will show in years to come. Not only do we have our own internal resources, but we look to third-party resources for detailed design or workload balancing around the world to not only our internal resources, but we can also draw on qualified third parties as we need to when we see ups and downs in the demand for projects. Finally, Ray mentioned and I really wanted to emphasize that our philosophy around product lines and product line platforms allows us clearly to leverage our engineering capability, minimize the additional or repeat engineering we might have to do for that core plant, spend our energies in the engineering teams for the what's required on the specific project requirements around the world. When we look at a team that's global, we have to make sure that they're mobile. And there's 2 ways to make a team mobile. One is with information technology, so they don't have to get in an airplane with a roll of drawings. I've been around 30 years, it used where you'd be looking at a plane and people were carrying lots of stuff. That's all electronic now, we have common platforms, so teams that are designing in China can feed information to Korea. People who are in Mexico can feed information up in the United States. They don't have to get out of their chair and get on an airplane to do work for another country or another region to support this network of global engineering. But when they need to, they're also mobile. I have engineers from the Russia team in the Shanghai office today working on those Russian projects because that engineering is being done in China and both for the ownership of that project, but also for the training and development it's important for them to go to Shanghai and really work with the team there. And likewise, around the world, I have people moving around as they need to because of expertise or because of training, they really make sure that the job gets done and they go where they need to go. So finally it really is a recognition that we have this global product line approach. But we have local execution teams that really deliver in the local culture, local language and local capabilities.
Finally, it is all about results, so the last slide is really my favorite slide. I think many companies in the capital-intensive industries that are out there, not just industrial gases, would be hard-pressed to put this kind of slide together. You can see over many years, we have a track record of first-class, world-class results in costs and schedule, why do we know this is world-class? As Steve mentioned this morning, he hinted to this concept of first quintile performance, with that independent project analysis, IPA runs benchmarking that we regularly participate in and they tell us as they compare to other industries who have these embedded engineering organizations, that we're performing in the top tier of those organizations and regularly doing that. And that's hard to do. Very challenging to hit this kind of numbers as you can imagine. This is average cost of portfolio. We have challenges in a couple places and with a few projects, but on a whole, this is recognized as world-class performance.
Just to speak to a few of those challenges which are highlighted in the bottom 2 blocks. Firstly, to Russia. We talked a lot about Russia and Todd hinted that, that is a challenging place to do business. Todd, Scott, myself, Rey, have been out to the job sites in Russia and apart from the weather, it's just a big place and the challenges of moving things around and getting things done are very, very difficult. However, we've applied a few solutions. One of those is prefabrication. It sounds like an easy word, but one of the things we've also done is look to prefabricate in China. To give you an example, EVRAZ's project that was on Todd's map, we moved close to 20,000 tons of material and equipment from the port of Antwerp and the Port of Shanghai, through the Black Sea, up the river and overland into that job site rather than looking at the supply network and supply chain in Russia. We sourced the material from key suppliers in Europe and China and then relied on the local team to execute and finish up that project, took risk out of the field, allowed us to manage the low-cost and productivity we see in China suppliers. Having said that, it is critical as I mentioned to make sure we get the local talent because in the end, a large piece of our project costs are about construction costs whether that's in China or Russia or Brazil, understanding the local market is critical, making sure we manage that well with the local team is critical. Contractor performance is difficult for us. It's difficult for lots of companies out there as trades, the load on trades gets very much higher as more and more projects get approved. And so worldwide, there is this challenge with finding good contractors, good tradesmen to get out there and do the work that's required. Finally, another example about Brazil. With the World Cup and the Olympics, we are seeing challenges in finding contract availability in Brazil. We are leveraging our factory in Brazil historically that made cold boxes and other components for us. What we've done is look to them to expand their capacity, to actually operate as a contractor and provide field work for our projects rather than going after a third-party so we bring that in our control, even though it's fieldwork, but we're managing ourselves to do a lot of the contracting work in Brazil. So finally, the message is that this proven execution capability is all about making sure we're delivering the kind of expectations that are given to us by the organization and ensuring we're delivering the project returns that Steve and the senior management signs up for. So with that, I'll turn it over to our Chief Financial Officer, James Sawyer who will close the program.
First of all, I just like to thank you all for choosing to spend the day with us today. It's very important for us to have our best analysts and best shareholders in the room to get to get together with our best internal talent. So thank you all for coming and spending the day. We're going to do another Q&A session as soon as I finish speaking, but I thought the most important question in front of the house and I think has been for a while and for myself, is can we continue to deliver the same kind of performance over the next 10 or 20 years that we did over the last 20 years? So with that, I thought I just put up this chart here, which shows our compound annual sales growth since 1993, 8%, a little bit by coincidence, but a little bit by intent, that's what we said we can do going forward, but clearly, that's a number that is more than 2x what industrial production has been over that 20-year time period. So back in the 90s, people have said, "Well we can grow, maybe 1.5x industrial production," but a combination of a little tailwind from industrial production, application technology that Rey talked about and John Paniker talked about plus project backlog, plus maybe 1% from acquisition this is about how it got us to 8%. Sales growth and I think based on what you heard today, I'm certainly convinced that we can continue to do that on average going forward. Now we also managed to leverage all the 8% sales growth to 11% operating profit growth on a compound annual basis and thereby increasing the operating margin year-by-year as time went on. I think that our practices in productivity and pricing are the main drivers of that, offset a little bit by inflation, but also the large backlogs on site projects will tend to come in at a bit higher operating margin percentage than the base business because those are the dimensions of on-site, low capital turn over, but high margin. And lastly, EPS growth over 20 years has averaged 13%. So we have been getting leverage from operating profit down the income statement through basically running shift with a higher return on capital because high return on capital by definition means that each year you're generating a lot of cash flow out of the assets that you've got. If you can generate more cash flow from the assets you've got, you got cash flow to reinvest in the business and more cash flow to pay out dividends, in buying back stock and that's where we get that extra bit of leverage. So I thought this chart kind of summarize how we fit everything together of what you've heard today in terms of meeting these goals. We're talking about high single digit sales growth. Operating margin expansion and double-digit EPS growth. And just to recap a little bit, how we've talked over the last 5 or so years. We've talked about what we call the 3 Es, energy, emerging markets and environment as being somewhat secular growth drivers above and beyond what you might expect to get out of normal industrial productivity growth and I think some of the questions over the last year is that the bloom off the rose in emerging markets? Well, yes, the last 2 years have not been the greatest years for emerging markets, let's put it that way, but my belief, any people's belief is that, as these emerging countries improve their standard of living over time, the demand for industrial gas is going to go up and we're very well-positioned to take advantage of that. So you heard Todd talking about our projects in Russia, Domingos in Brazil and the infrastructure builds in Brazil, Antonio Cesar in Mexico and PEMEX's challenges in industrial oil and gas and Anne Roby talking about China gas gasification which, of course, pulls in the energy element, but also about applications technology to help solve air and water pollution problems in China, in India, in the other emerging markets. And that, I think, is going to continue to be a strong, strong secular growth driver.
Now on the energy side, I think if I flip back 5 years ago, I think before the advent of shale gas and horizontal drilling, we all would have expected that sometime in the future, the U.S. would become more dependent on energy from tar sands and heavy oil and sulfur-rich crude feedstocks, which was one of the big drivers of the hydrogen projects that we saw on the horizon. Clearly, a lot of those have been building, clearly a lot more are going to be drilled, probably more for hydrocracking and for hydrotreating, but if you put it all together, there's probably going to be a few -- fewer hydrogen plants built, not only in North America, but globally than we thought a few years ago. Now, is that going to be big change? I don't see that because I think it's more than made up for by the U.S. petchem projects and the kind of industrial Renaissance in the United States, but Canada's continuing to grow nicely. Tar sands projects, which we all kind of gave up for dead a few years ago have come back with the red water project and there are going to continue to be other projects like that and we're going to keep growing applications in intensity. I think Rey put up a very convincing chart of the amount of oxygen consumed per ton of steel production continues to go up year in and year out and we have similar charts for welding, chemicals, most of our end markets where the intensity of our product used per ton or unit of our customers product continues to go up at a pretty rapid pace. Scott talked about the high performance culture, Scott, thank you for the horses and Scott Kaltrider here, I think, talked about disciplined pricing. Now if I was a customer and Scott was a salesperson, he told me he we're going to have price increase, I'll say, "Okay, Scott, I take it, coming along." Scott is a very convincing person on that and he convinces not only our customers, but what's more important is to convince your sales people in your organization to have the fortitude to increase price. We talked about integrated supply and driving for density. Eduardo talked about how we structure the business to take advantage on the one hand of getting density, of the advantages of integrated supply between on-site merchant and package, but at the same time, trying to drive the behavior so that we have relatively dimensioned P&Ls. And so people have a P&L, they know how to drive it within that whole system. Energy management, one of our biggest cross items. I think we're probably one of the best electricity purchasers in the world in terms of a 12 person department who is always on top of managing our energy contracts, daytime, nighttime energy. It's a big driver of productivity and a big driver of improving margin.
And then lastly, cash flow. We have industry-leading return on capital. It's been somewhat subdued from 15% down to 14% to mid-13%, primarily because of the recession and because of the high amount of project backlog that we have in projects under construction today. As those projects come on stream and we got a little bit more capacity utilization, we should be getting back to the 15% area. Now the other area dear to my heart is cash flow and what are we going to do with all that wonderful cash flow generated from operations. As you can see over the next 5 years, we expect to generate about $17 billion in operating cash flow before capital spending acquisitions or dividends. It's roughly about 24% of sales, I keep pushing to 25%, maybe we'll get there some day, but it's a very strong number of operating cash flow from sales. We're planning to keep the capital spending in probably $1.8 billion to $2 billion range, which albeit a smaller number than we spent last year, but very well tied in with what we see as strong, high-quality projects. The kind of projects that Ben Glazer talked about when he did his whole presentation on project analysis. And as you can see, more of that is coming from the North America and being kind of reallocated a little bit more to Asia. Europe stays constant, but that's basically kind of a reallocation of cash flow from southern Europe to reinvestment in Russia. And South America stays relatively constant as well.
That end, according to my model, should give us enough cash flow to still spend about 15% of sales and CapEx and still deliver an over $7 billion of over 5 years to a combination of increasing dividends and cash used to purchase stock, probably to the tune increasing from maybe 1% to 2% of the shares outstanding. And so that's kind of the finishing remarks that I have. Now I'd like to turn it over to some question and answers.
Question on the -- clearly ROIC is a very important metric for all of you, and you touched upon some of the hurdle rates, 500 basis points over internal, I don't know, some threshold. Could you guide us into the environment that you confront over the next 10, 15, 20 years compared to the past 20 when interest were declining versus every expectations suggests they should go up over that next secular cycle. How does that play into your assumption? And are you -- when you have a take-or-pay contracts, the inflation index in the developed world, which I know you do manage to get in the developing world.
Stephen F. Angel
I'll take. First of all, everything is a U.S. dollar-based return. So when we're looking at a project in India, we're looking at many things, one of which is, what do expect from inflation? And therefore, we expect high inflation coverage in our contracts. What do we think is going to happen with respect to currency depreciation over the years? We look what the rupee has done historically. We make an assumption about that going forward. So everything is brought back to the U.S. dollar-based return, all projects, let me just start with that. And we've been saying that the minimum criteria, the minimum hurdle rate for internal rate of return is 500 basis points over the cost of capital. So 500 -- and then we go through the adjustments that Ben talked about, many cases, we may go in and say we need to have at least 16% on this project or even higher. And again, it depends on how things fall out, but the minimum would be 500 points above the cost of capital. That's how we think about it. Yes, interest rates have fallen, they've fallen in recent years. But we've really seen very little difference in the marketplace in terms of project competitiveness. And one thing I'll remind you of is -- and I made this point many times. When we look at the playing field of the projects that we participate in, about 25% of them, there's no competition. And that's one of the benefits of a density strategy. We're working with customers and they're continuing to grow, we're growing right along beside them, so no competition, another 25%, 1 competitor next 25% would have 2, and then about 25% of the time all 4 majors are going after a piece of business. So it's not as competitively -- it's not as -- it doesn't have a competitive intensity that you might think where all 4 of us are going after the same thing. It's just who's going to take the lowest return. It happens maybe a quarter of the time. So I don't think going forward, and I can always find an example here or there, where I can say somebody broke ranks, but I could say that a year ago, 3 years ago, 5 years ago, 10 years ago, that's really unchanged. But I think what I describe is going hold true going forward over the next 5 years or so. I think we can maintain that discipline around returns and I think we can win the projects we want to win and we can execute them the way we've been executing those projects.
Go back here.
John E. Roberts - The Buckingham Research Group Incorporated
John Roberts, UBS. Jim, before you run away, if you look at Slide 66, which is the sample project return on capital slide that's there, obviously the whole business is not projects. So about -- between year 5 and year 6, you crossed over the corporate average and the area to the right of the breakeven is much, much larger than the area left. If the whole business were projects, the average return would be much, much higher than the 12% to 14% range. Could you just conceptually talk about the bridge. Is it just subtracting out the SG&A and R&D, which is the drag on returns, or when projects get too old, do they come down very rapidly in the late years and there's a mix of some really old projects. How do you delve down from a weighted average of this project return, which is really high down to the corporate average?
Stephen F. Angel
Jim, the question really was about how do we relate return on capital and the project cycles and the startups and then further down the road and in fact, that the entire business is not projects, it's a merchant package, why don't you just relate that.
James S. Sawyer
That is a long question, but the first thing I would say is that we basically evaluate decisions on a cash flow basis, primarily using internal rate of return, which is discounted cash flow over time, rather than just looking at the annual static return on capital and I think because the annual static return on capital is really just kind of a snapshot in time, it really can't take into consideration the cash flows going forward. So we're generally targeting a number around 13%, 14%, 15% after tax IRR. Some of the smaller projects in our own backyard or higher than that. But when we evaluate a project, we look at a number measures, and one of the ones that I look at, kind of takes out the effect of depreciation, which is I look at with the EBITDA divided by the undepreciated capital of the project and I like to see that get to about 20% within 2 or 3 years as kind of a number to look at there. So I think the book return on capital is something that you look at after the fact and not so much in the decision-making process. Now in terms of the merchant and the rest of the business, that's really where you can climb up on the return on capital because if you start out with a base load project with a large on-site customer, I'll make it up, say, it's a 12% IRR, over time starting at a couple of years, and maybe even more important in 5 or 10 years, if you have a strong merchant business density, and a lot of integration between your on-site merchant and package, you should be able to move that up to a much higher return on capital on a year-by-year basis.
Stephen F. Angel
Come over here to Mark Gulley.
Mark R. Gulley - BGC Partners, Inc., Research Division
Mark Gulley, BGC Financial. In the use of cash flow slide, just 50-50 stock buybacks and dividends. No provision for acquisitions. Can you talk about the role of acquisitions going forward and is it just a rounding error, it's in the cash that you're . . .
Stephen F. Angel
Probably just more of a rounding error. I mean it's -- I think we'll spend maybe $100 million to $200 million, not a significant amount, Mark. If I don't spend that amount of money, it doesn't bother me. It's just going to be the quality of opportunities, but these tend -- these are intended to be smaller kind of U.S. packaged gas opportunity to the U.S. We may see a few things around the world globally like I said earlier, nothing like what we with UC02. I can be standing here 10 years from now. I doubt I will, but I could be standing here 10 years from now and nothing will have happened of that magnitude.
You go in the back there.
Michael J. Harrison - First Analysis Securities Corporation, Research Division
Mike Harrison with First Analysis. Anne showed us the map of China that kind of showed you're located mostly on the coast and have very good density, and then there was a comment in there that the inland portion of China is expected to grow more quickly than the remainder right now. Can you just kind of reconcile that for us and talk in a little more detail about how you approach China? Does that maybe explain why you guys have a relatively smaller portion of your backlog based in China?
Stephen F. Angel
Anne, why don't you just take that question?
Anne K. Roby
On the map, we have an investment that's underway in Chongqing, which is one of the hotbeds of inland activity. And we are also looking at some other areas to set a new footprint and add some density, but very selectively, right, because there are a lot of areas inland that are under development, but will not end up being the next Shanghai or next Beijing, but Chongqing looks very promising.
Stephen F. Angel
Question over here.
Vincent Andrews - Morgan Stanley, Research Division
Vincent Andrews from Morgan Stanley. A lot of conversation today about contracting and pricing and pushing energy through, and then one of the risks you mentioned to your forecast is actually foreign exchange. So I guess my question is, how come foreign exchange is not one of the components of the contracting process or is that something that might change in the coming years now that it seems like the dollar is going to adopt going in your favor?
Stephen F. Angel
Well we -- the question is on foreign exchange. We do have some contracts where we have been able to tie the return to a U.S. dollar return. It's indexed to a U.S. dollar return upfront. So obviously that can be preferential in terms of risk mitigation tool. But where we haven't been able to do that, there can be reasons, various reasons for that, we want to make sure that we've got all the inflation. Inflation covered 100%, so if you're on a high inflation country, and you don't have high inflation coverage, you're going to have a problem later on in the return of that contract and like I said earlier, we tend to be very conservative about where we think currency is going to be heading. I think in every currency of the world we look at, we apply some depreciation factor in it. So in other words, we assume every project around the world whether you think the China RMB is going to strengthen or not, will have some erosion, some depreciation factor into RMB, just to be conservative because we know this is a 15, 20-year asset and we just don't want to get that wrong. But we do on occasion, where we can, we'll tie the formula upfront to a U.S. dollar return, but if not, we try to make some very reasonable conservative assumptions.
Stephen F. Angel
Pardon me? Will you just say one more? Okay. Kelcey's in charge.
Kelcey E. Hoyt
All right. This gentleman right here beside Mr. Gulley.
You mentioned that the currency is factored in, why do you not hedge the exposure to some of the currencies if you think these currencies would be a major factor for you?
Stephen F. Angel
Jim, question is about hedging relating to foreign currency, why don't you take that.
James S. Sawyer
Yes. Well, I think that we're no smarter than the rest of world about which way currencies are going and we try to denominate some long-term liabilities in local currency debt, for example, or index contracts back to the dollar, but I think the best way for us to manage our currency exposure, if you will, is to be transparent about it because with the accounting rules, you can't hedge more than 1 or 2 or 3 quarters out without having to mark-to-market your hedged gains and losses and then that gets all mixed up in your operating results and your financial results. So each quarter, when we give earnings guidance, we say what the assumptions are, what the current spot rates are for that quarter and the currencies change over time. People can adjust for that in their own forecast, but in my mind, people should be investing in Praxair for its operations and its operation capability in different countries and products and markets and go on with your eyes open about how much our earnings are coming from the euro, Canada, Mexico, in Peru and Brazil and so forth.
Stephen F. Angel
Very good. Listen, I want to thank everyone like Jim. I want to take everyone for coming today. I know it's been a long day. There will be a few people hanging around if you had a question, a burning question that you didn't have a chance to ask and have answered. There will be a few people around to do that and, everyone, have a safe day. Thank you.
Copyright policy: All transcripts on this site are the copyright of Seeking Alpha. However, we view them as an important resource for bloggers and journalists, and are excited to contribute to the democratization of financial information on the Internet. (Until now investors have had to pay thousands of dollars in subscription fees for transcripts.) So our reproduction policy is as follows: You may quote up to 400 words of any transcript on the condition that you attribute the transcript to Seeking Alpha and either link to the original transcript or to www.SeekingAlpha.com. All other use is prohibited.
THE INFORMATION CONTAINED HERE IS A TEXTUAL REPRESENTATION OF THE APPLICABLE COMPANY'S CONFERENCE CALL, CONFERENCE PRESENTATION OR OTHER AUDIO PRESENTATION, AND WHILE EFFORTS ARE MADE TO PROVIDE AN ACCURATE TRANSCRIPTION, THERE MAY BE MATERIAL ERRORS, OMISSIONS, OR INACCURACIES IN THE REPORTING OF THE SUBSTANCE OF THE AUDIO PRESENTATIONS. IN NO WAY DOES SEEKING ALPHA ASSUME ANY RESPONSIBILITY FOR ANY INVESTMENT OR OTHER DECISIONS MADE BASED UPON THE INFORMATION PROVIDED ON THIS WEB SITE OR IN ANY TRANSCRIPT. USERS ARE ADVISED TO REVIEW THE APPLICABLE COMPANY'S AUDIO PRESENTATION ITSELF AND THE APPLICABLE COMPANY'S SEC FILINGS BEFORE MAKING ANY INVESTMENT OR OTHER DECISIONS.
If you have any additional questions about our online transcripts, please contact us at: email@example.com. Thank you!