There Is No Inflation: Too Much Supply, Not Enough Unions

by: Saad Filali

Recent data suggests inflation has peaked in the summer for the short term.

Globalization, oversupply and online retailing will keep pressuring prices down.

The Phillips curve died with unions.

Inflation will remain subdued for a long time hence elongating the business cycle.

Data Review

Source: FRED via BLS

August produced a slew of data that surprised to the dovish side with the highlight being a miss in CPI by 0.2% - a pretty big miss as far this particular data release is concerned

So, what did the consensus miss exactly? We need to check out the details that often get overlooked in financial media.

Source: BLS

There are a few things that stand out here, first all in all, most categories are very tame. Inflation isn't generalized. Many key categories are actually still negative on a yoy basis. Let's discuss some of the most interesting ones:

Food: As seen above, food inflation is very weak. If it wasn't for transportation and energy prices, there would be none at all.

From HRL's Q2 transcript

Our balanced business model allowed us to mitigate double-digit increases in freight, a dynamic park environment and continued oversupply in the turkey industry.

Low food inflation is a global trend that is not going away. It is mainly explained by huge investments that were made during the previous cycle and ever-increasing yield/efficiency in agriculture and the food industry generally.

Source: FAO

Consider also that the health and wellness trends in food are leading consumers to reject some commodities and lower consumptions overall. These trends along with some supply issues have plunged sugar in a terrible bear market for example.

Source: FAO, Proprietary work

The trade war and oversupply have sent many US agribusinesses into dire financial stress. Below an excerpt from Tyson Foods' (NYSE:TSN) latest earnings addressing both subjects:

Thanks Jon, and good morning, everybody. Thanks for joining us today. While our business continued to grow in the third quarter, we are clearly not satisfied with our results, particularly in our Chicken segment. Intertwined with uncertainty in trade policies and tariffs are increasing supplies of relatively low-priced beef and pork that are competing with chicken.

Rising freight costs have been a challenge for all of our businesses. We now expect freight to be about $270 million more this year compared to last year, with a net effect for FY 2018 estimated to be around $0.33 per share.

The trend in food prices is likely not going to reverse meaningfully outside of momentary reversals due to supply constraints and better management of stocks that never last that long (see wheat futures recently). Leverage has also been increasing among US farmers so these folks can't even cut production to save themselves.

Source: USDA via Deere & Co

Source: USDA via Deere & Co

Another category that is contributing to outright deflation is apparel. I could go on and on with data from retailers, but all you need to do is check stocks of retailers with second class or weak online exposure. The price filter is the single best thing that ever happened to consumer purchasing power since globalization itself. I cannot possibly overstate the power this filter has on prices. It's gotten to the point where you have price comparators of price comparators for every single thing you can think of: transportation, apparel, insurance, medical services, services and soon, if we believe AMZN, fresh food.


Energy is the only red spot on the map and that's because oil benefits from its affiliated cartel. Let's remember that around a year ago OPEC was in crisis mode with oil consolidating below 40$. If it wasn't for OPEC forget inflation, the world would be facing deflation.

  • Their upside has been capped by President Trump himself. In a tweet, he clearly told OPEC & friends to limit oil upside lest he get involved with them personally.
  • One could think that actually worked but that's not entirely true. OPEC are aware that oil drives the business cycle as much as credit and that if they went overboard with it then they would cause a recession just like they did in 2014/2015. Ending up with very low prices that are unsustainable (for them) once more. If Iran/Venezuela situation improves or is permanently fixed, then I predict oil prices will converge towards 50.
  • Point is I don't see where the upside for oil would come outside of another conflict in the ME potentially caused by Hezbollah. It is difficult to assess the appetite or the social/political capital of Tehran to engage in such a project, even though they'd love to, because the economic situation in Iran is really difficult and the population is looking towards improving their own situation first.
  • This is a very different layout then what we had in Turkey which was actually doing really well before Erdogan went rogue.

Inflation expectations

Expected inflation data can be segregated into three categories: Surveys, Hard data and Market data.

I have found surveys, such as the Michigan U inflation expectation survey, to be useless because layman inflation expectation is mostly back forming. Meaning they just look at what it has been and think it's going to continue.

Hard data type such as the PPI are useful because they form an essential input into the models of the analysts who contribute to the consensus and thus will be market moving if we get enough of a surprise.

Market driven: real rate/breakeven proxies thus the yield on TIPS or forwards (they are different, the spread between the two corresponds to an inflation uncertainty premium but this is beyond the scope of this article).

Real rates have mostly been driven by the FED and oil, the only meaningful source of inflation as we saw

Source: TradingView TIP vs TLT and 50d rolling correlation between TIP and USOIL (WTI)

Thus as long as oil is range bound and the FED follows a scenario that I previously outlined here (all of the trades I recommended have been working very well, especially the Nikkei225/EWJ which I strongly recommended twice recently), I expect these trends to continue for now.

Wage inflation: Unions matter

So many articles and equity research notes have been talking about wage inflation. Fed voters, staffers, bank strategists and hedge funds analysts say that they do not understand why the Phillips curve is not working. They also always add that it is going to work someday in the near future. I argue that post crisis corporate management and the demise of unions have killed the Phillips curve. The idea is seducing: most people who want and can hold a job are employed, thus further employment gains can only occur with wage inflation that leads to general price inflation. Except, the relationship and dynamics between employers and employees is not simple. Cost cutting is the mantra of MBA programs where corporate leaders stem from. They are taught that raising wages is unsustainable, employee productivity gains decline marginally with age and experience thus it makes no sense to raise wages besides acquiring the personnel in the first place. Indeed, the easiest way to get a pay increase these days is to switch companies.

Source: Nomura, ADP, ATLANTA FED

Managers are still painfully aware of the ramification of the business cycle. They have learned that you shouldn't binge spend on capex just because the times are good now. They believe that good times are the precursor of bad times. Little do they know that their sensitivity to this mean reversion in business conditions is slowing the cycle, making it actually useful to invest in certain industries where the oversupply from the previous cycle has been cleared. There is a reason why growth is so popular and this is one of them: very few people are actually doing it even though that is improving today if the past few earnings seasons are anything to go by.

I sincerely believe the reason there is such a strong belief in the Phillips curve in the first place among economists (especially of the bank strategist type) is because it's been true when they were growing up and the first thing they are taught in college. It's a first-year econ 101 staple. The problem with that is they don't look at the underlying math or assumptions and look at it as a direct application of the supply demand framework without questioning underlying conditions or dynamics.

The Phillips curve, at the root is a simple linear regression.

Source: OECD & IMF

No COLAs, no wage inflation

To me, it is clear that the 75-84 period is actually the outlier here and that a secular shift has occurred, it's up to us to explain it and to me the reason is very clear: the demise of unions. Can you go to your employer and demand a higher wage now? If you can, good for you. But most of us can't do that because we have no leverage, in fact the employer does. How then can employees get some leverage and get the employer to the negotiating table? By forming unions. And unions were very successful during the period when the Phillips framework actually worked. Unions had forced COLAs (Cost Of Living Adjustments) initially called Cost-of-living escalator clauses into the contracts of millions of Americans. Don't know them? Ask your dad or your grandpa. A bit of history is required here to understand that these clauses were the cause of the observed relationship between unemployment and inflation.

The first major COLA was between GM and the united automobile workers. By 1955, 23% of workers who had collective bargaining agreements also had COLAs. COLAs were very popular and a worthy goal for unions. During 76-81, 60% of workers covered by major union contracts also enjoyed COLAs. And they worked! During the 70s, wages of employees in heavily unionized industries grew significantly more relative to wages of other employees in the economy

Source: Journal of Labor Economics, Vol. 1, No. 3, (July 1983), pp. 215-245.

Can we have MAGA without MUGA?

Source: BLS, Proprietary work

The causality of what I said makes perfect sense but it doesn't mean the effects I'm describing were mathematically meaningful. That is why I will show with the following statistical study that unionization is indeed a significant driver of the steepness of the Phillips curve, aka the whole relationship between tight labor markets leading to wage inflation and then to inflation

Source: BLS, FRED, Proprietary study

First this chart shows us the coefficient of the Phillips curve (assuming a linear model) using the last 50 months at the date for each data point. This tells us what we already know, the Phillips curve worked in the 80s, 90s outside of recessions where inflation is driven by stronger economic forces. What it also shows is that this relationship has been indeed a lot weaker than in previous cycles, never regaining the highs above 2.5. staying stuck below 1. Perhaps even more importantly, this relationship is usually sinusoidal following the business cycle over similar periods, in this cycle the ramp period has been a lot longer. I believe this to be one of the causes of the length of the current cycle and argue that this cycle will be longer than anyone expects because of it. As I've shown in this article, the average hourly earnings variable is the most important for the business cycle.

Source: BLS, FRED, Proprietary study

Again, the same observation can be said about the supposedly much more direct relationship between AHE(Average Hourly Earnings) % change yoy and the unemployment gap (see article on Cycle Stage Determination)

Finally, we will study the relationship between unionization and the strength of this relationship (excludes recessions as per NBER)

Source: BLS, FRED, Proprietary study

Proving a satisfying relationship between unionization and the strength of the Phillips curve logic using data from 1983 to 2017. A Concave model tells us that some unionization greatly improves the steepness of the Phillips curve but also that the marginal increase in steepness is decreasing.


The demise of unionization, globalization and the surge in online retailing with price filters are all powerful structural forces that are not going anywhere and pushing inflation down. Food and Energy used to be the two volatile items that have contributed to the end of many bull markets but the former still suffers from oversupply, ever-increasing yields and global competition while the latter is caped by a more rational OPEC body. Subdued inflation has and will continue to contribute to an elongated business cycle characterized by capped rates nominal and real.

Recommendations: Long Equities, Long TIP, Long TLT, Buy LQD on dips

Enjoyed reading my article? Did you benefit from it? You can show your appreciation by following me here or on Twitter

Disclosure: I am/we are long TIP, LQD. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.