Seeking Alpha
Long only, growth, long-term horizon
Profile| Send Message|
( followers)  

© Elliott R. Morss

In my recent article on manufacturing, I pointed out that the vast majority of job losses in the last 50 years were the result of technology gains rather than foreign competition. I also predicted that the prospects for US manufacturing were quite good because the labor share of manufacturing costs is falling and there is a growing demand for Western goods from emerging market countries. In this article, the prospects for employment in other US sectors are explored.

"Tradable" Goods

Michael Spence just wrote an article making an interesting distinction between "tradable" and "non-tradable" economic activities. He argued that most job losses in recent decades have occurred in the tradable sectors - agriculture, energy, manufacturing, and mining. He claims these losses are primarily the result of labor saving technologies but also stem from "disintermediation - the elimination of intermediaries in banking, online retail, and a host of government services, to name just a few affected areas…."

Below, we look more closely at the tradable goods phenomenon and what is in store employment in other US sectors.

Tradable Goods - A Closer Look

Consider first US employment gains and losses since 1947. Table 1 indicates that since then, nearly all the job growth has taken place in service sectors rather than in goods production. 85% of all US jobs are now in the Service sector.

Table 1. - US Employment Growth, 1947 - 2011

Employment (thous.)

Compound Annual

1947

2011

Item

1947

2011

Growth %

Share

Share

Total

52,468

134,440

1.48%

100%

100%

Goods

25,453

20,430

-0.34%

49%

15%

Services

27,015

114,010

2.28%

51%

85%

Source: US Bureau of Labor Statistics

Table 2 gives employment change information in more detail with "tradable" sectors indicated. The information bears out Spence's contention that most job losses have been in the tradable sectors (under manufacturing, most durable goods as well as most mining and agriculture).

Table 2. - US Employment Losses by Sector, 1947-2011

Employment (thous.)

Compound Annual

1947

2011

Item

1947

2011

Growth %

Share

Share

Services

27,015

114,010

2.28%

51%

85%

Retail trade

4,390

14,643

1.90%

8%

11%

Health care

2,734

14,046

2.59%

5%

10%

Finance/Ins.

3,704

5,752

0.69%

7%

4%

Wholesale trade

2,171

5,529

1.51%

4%

4%

Education

443

3,241

3.16%

1%

2%

Manufacturing

14,428

11,733

-0.31%

27%

9%

Durable Goods

8,130

7,274

-0.16%

15%

5%

Computer/electronic products

5,782

1,107

-2.55%

11%

1%

Machinery

2,640

1,057

-1.42%

5%

1%

Motor vehicles and parts

4,328

717

-2.77%

8%

1%

Nondurable goods

6,627

4,460

-0.52%

13%

3%

Food manufacturing

1,613

1,456

-0.16%

3%

1%

Construction

2,130

5,504

0.67%

4%

4%

Information

1,659

2,659

0.74%

3%

2%

Mining and logging

1,005

784

-0.34%

2%

1%

Coal Mining

266

86

-1.75%

1%

0%

Oil/Gas Extraction

254

174

-0.59%

0%

0%

Utilities

481

555

0.07%

1%

0%

Farming

7,890

2,254

-1.94%

15%

2%

Source: US Bureau of Labor Statistics

Dramatic job losses have occurred in all tradable sectors with the largest happening in farming, computing and motor vehicles. In contrast, employment in all Service sectors has grown since 1947. In the following section, the reasons for these job losses are examined.

Productivity

Table 3 provides labor saving productivity data for a broad measure of the US economy (Business)1 and Manufacturing. The "Unit Increase" statistic indicates how many more units of any product a laborer can produce than in the earlier year. For example, the 5.5 number for business in the 1947-2011 column means that if a worker could produce 1 unit in 1947, that worker could produce 5.5 units today. Note the striking difference between productivity gains overall (Business) and the much larger ones in Manufacturing.

Table 3. - US Labor Productivity Gains

Unit Increase per labor hour

Item

1987-2011

1947-2011

Business

1.9

5.5

Manufacturing

3.1

20.0

- Durable

3.4

26.3

- Non-Durable

2.7

14.0

Source: US Bureau of Labor Statistics

Table 4 gives productivity gains for selected manufacturing sectors. Since the data for these specific items do not go back to 1947, I have assumed they have increased by the same amount as manufacturing overall.

Table 4. - Labor Productivity Gains, Selected Manufacturing Sectors

Unit Increase per labor hour

Item

1987-2011

1947-2011*

Computer/electronic products

11.4

73.1

Automobiles

2.2

17.2

Machinery

0.7

4.7

Source: US Bureau of Labor Statistics

* Estimated assuming increases same as increases for Durable Manufacturing.

The productivity gains in the computer sector are amazing. But the gains in autos are also extremely significant. In essence the data indicate that if it took 10 men on the assembly line to produce 10 cars per day in 1947, 10 men could produce 172 cars per day in 2011. Or turn this around using the same assumptions: in 2011, it would only take 58% of one worker's day to produce 10 cars. We have heard a lot of moans and groans about out-sourcing of jobs to low-cost labor overseas. At least for autos, this has not happened. Most of the jobs that have been lost in autos have gone to Japanese and German producers who have high labor costs and are as proficient as Americans in designing mass production systems.

Location Effects

Spence offers the following on how the new technologies will affect where goods are produced:

"The same class of information technologies that automate, disintermediate, and reduce the costs of remoteness are also enabling the construction of increasingly complex and geographically diverse global supply chains and networks…. Global supply chains - constantly in flux, owing to rising developing-country incomes and shifting comparative advantage - locate productive activities where human and other resources make those activities competitive. The result is what is sometimes called the "atomization" of global supply chains: increasingly fine subdivisions are feasible, more efficient, and locatable almost anywhere."

On this point, I have my doubts. Sure, complex supply chains and networks allow products to be produced just about anywhere, but I believe the pendulum is swinging in the other direction. As I argued in a recent piece, global companies are not as enamored with "increasingly complex and geographically diverse global supply chains and networks" as they have been in the past.

There are several reasons for this:

  • The Fukushima disaster stopped much Japanese production for several months - a real problem for complex supply chains sourcing in Japan.
  • The global warming projections of more severe weather are also raising questions about the need for long supply chains;
  • Corporations are increasingly concerned about their global CO2 footprints. Air transport increases these footprints in a hurry.
  • Comparative advantage and competitiveness are fine when the world is at full employment. However, in the aftermath of the global recession, countries are looking for jobs and are taking steps to get them by limiting access to their markets. It is no surprise that so many parts for Boeing's 787 are produced in Japan: the two Japanese airlines have orders for 111 of them, far more than any other country.

Table 5 presents productivity data for major Service sectors. For these sectors, "Units" are defined by value added rather than by number of units produced. And here also, the productivity gains in banking and trade have also been impressive - a clear manifestation of the disintermediation" discussed by Spence. Coming up with accurate productivity measures for government, health, education, and other Service sectors is problematic.

Table 5. - Productivity Measures for Selected Service Sectors

Unit Increase per labor hour

Item

1987-2011

Commercial banking

2.4

Wholesale trade

2.1

Retail trade

1.9

Postal service

1.2

Source: US Bureau of Labor Statistics

Tradable and Service Sectors - Looking Ahead

To understand what is likely to happen in the next ten years, it is worth reviewing what we know about technological change. This is relevant because a major technological revolution is underway - the information revolution. I was privileged to work with a group of historians on this subject for more than a decade. So what did I learn? What does history tell us about technological revolutions?

  • It is hard to project what will happen and large sums of money will be made and lost in the process;
  • The results are not always the best outcome: sometimes up-front marketing monies and government lobbying efforts dictate the outcome;
  • And even when it is clear what the technologies can do, getting humans to change their behavior to take advantage of new technologies takes a long time.

So how do these lessons apply to the current information revolution? So far, we have the Internet and the ability to substitute machines for humans in production. And we applied this new knowledge in the tradable sectors: rote labor was replaced by machines - the image of one worker looking at a computer monitor in a factory with no other workers is apt. And the same thing happened on farms where today one farm worker manages an average of 137 acres. But when it comes to improved information processing, things are just starting, there is far more to come.

The financial sector is accomplishing a volume of financial transactions that would have been unimaginable 20 years ago. Humans have been slow to adapt. Most people still pay most of their bills with cash or checks. In ten years, most people will pay bills via electronic processing over the Internet.

Retail sales: in 2011, Internet sales were $256 billion (growing rapidly), more than department store sales ($184 billion), and 15% of total US sales. Where will we be in 10 years? It is easy to imagine the Internet share exceeding 50%.

Consider health delivery: The Internet gives users the potential to diagnose their own symptoms - few use it. That will change. Individual health records, mostly printed on paper, get lost. That will also change so patients can have their records at all times. And there are major changes coming in the new health care bill to reduce health care costs by reducing supply-driven costs and delegating more authority for diagnosis to nurses and nurse practitioners. We know how to do all this. The information revolution has made it possible. The only question is whether the stakeholders' lobbies (hospitals, doctors, etc.) will keep it from happening.

Education is still mostly in the classroom with a teacher for each class. That will change as Internet education gains a foothold. Tom Friedman has been covering the rapid developments in this sector.

Summary

In short, what am I saying? Applications of the Information Revolution first took place in the tradable sectors. But major additional applications are coming in the Service sectors. Investment implications: I would not buy REITs that focus on shopping malls such as CBL & Associates Properties (CBL), Developers Diversified Realty (DDR), Federal Realty Investment Trust (FRT), Glimcher Realty Trust (GRT), or Kimco Realty Corporation (KIM).

Employment Implications

Much of the technology discussed above will be labor saving and require a larger portion of the labor force to be well-educated with considerable technical expertise. Michael

Spence ends his piece with the following:

"…it is possible that we are entering a period in which major adaptations in employment models, work weeks, contract labor, minimum wages, and the delivery of essential public services will be needed in order to maintain social cohesion and uphold the core values of equity and intergenerational mobility."

I agree. And as I have suggested in an earlier piece, the number of hours weekly hours worked continues to fall both globally and in the US, and this should continue. And more incentives for people to work less would help (like reducing the age at which workers qualify for Social Security payments.

On US labor, there is one more point worth noting: 9 to 5 jobs with a 40 hour work week are on the decline. Part of this was the result of the global recession and companies not wanting to pay fringe benefits. But there is another force at work - the Information Revolution. Workers are better educated than ever before and have access to all work information wherever they are. Why sit in an office? People are increasingly choosing to work part time. There is a growing demand for temporary work, and this happened even before the recession. Between 1990 and 2007, employment at temporary job employment services increased 126%, and it continues to grow.

Companies providing part time jobs should continue to grow - Manpower (MAN), Kelly Services (KELYA), and Spherion (SEN).

1 The Business sector is a subset of the domestic economy and excludes the economic activities of the following: general government, private households, and nonprofit organizations serving individuals. The business sector accounted for about 78 percent of the value of gross domestic product (GDP) in 2000.

Source: Tradable Goods And Employment Futures