Friday, January 30, 2009

January 2009 Early Indications: The Job Issue

Writing just days after roughly 60,000 layoffs were announced, it's difficult to look anywhere else for stories to analyze. Many facets emerge as one studies the employment question, and some historical context provides some surprises.


It used to be said that the bigger they are the harder they fall. More recently, industry consolidation was partially justified by both "synergy" and economies of scale. The result was companies that may have been too big to manage: Ronald Coase's theory of the firm implies that when the costs of bureaucracy limit market responsiveness, it's time to scale down. While some firms are currently said to be "too big to fail," I think we will see proof to the contrary relatively soon.

Where will all the new jobs come from? Mass hirings are infrequent even in the best of times: "GM adds 4,00 new machinists" wasn't something one saw in the news, regardless of the era. Jobs get added far more slowly and in more dispersed fashion than they get cut, especially when some firms are measuring severance in the tens of thousands. Part of the policy challenge is the asymmetry between the big cuts and the reality that small, growing firms add jobs by the handful or dozen.

The Biggest Employer

One key element of the story is the role of government as a direct employer even before stimulus-related jobs are counted. Ever since World War II, the federal government has been increasing as an employer, either directly or indirectly. For example, seven new cabinet departments (plus the EPA, founded in 1970) are less than 50 years old and reflect the growing scope of government:

Housing and Urban Development (1966)
Transportation (1966)
Energy (1977)
Health and Human Services (1979)*
Education (1979)*
Veterans Affairs (1989)
Homeland Security (2003)

*Broken out of Health Education and Welfare

Headcount at these agencies is significant: the Department of Agriculture employs about 85,000 people, making it a) the biggest federal organization not involved with domestic (DoJ, DHS) or foreign peacekeeping and b) about half as big as Cargill, a $120-billion food processor.

As of Q1 2008, the biggest employer in Pennsylvania was the State of Pennsylvania (no figures were released in the document, compiled by the state Center for Workforce Information & Analysis). #2 was the U.S. government, even after the closing of the Philadelphia naval yard in 1995. As private sector employment and profits drop for the foreseeable future, how will public-sector employers maintain their payrolls? The state of California, ahead of the curve in some matters, may be the bellwether here, and the story looks grim as the shortfall through 2010 reaches $40 billion.

Information-age Stimulus

Borrowing from the title of a recent newsletter, I want to return to the question of how a government stimulates a services-driven, information-centric economy. As a state Pennsylvania is reasonably representative, with a relatively large economy (#6 out of 50 states), and a per capita income almost perfectly at the median, ranking 26th of 50. Agriculture is important but not predominant, and 25 Fortune 500 companies are headquartered here.

To continue that list of Pennsylvania's top employers, note the paucity of private-sector job-generators:

3) Wal-Mart

4) City of Philadelphia

5) University of Pennsylvania (roughly 35,000 jobs, including a big medical center)

6) Philadelphia school district

7) Penn State University (not counting the affiliated medical center)

8) Giant Food Stores

9) UPS

10) University of Pittsburgh Medical Center

11) University of Pittsburgh

12) Weis Supermarkets

13) State System of Higher Education (public colleges and universities excluding Penn State, Pitt, and Temple)

All told, 20 of the top 50 employers in Pennsylvania are not businesses in the traditional sense of the word: that's 40% of the leader board, including six of the top seven. Half of the 30 largest private-sector employers, including eight of the top 12, are retailers, known for relatively low wages and, according to the BLS, the highest turnover among major sectors. As Circuit City demonstrated, they are also sensitive to economic downturns and could themselves be at further risk.

More significantly, Pennsylvania is officially a services economy: only one employer (Merck) in the top 25 and four in the top 50 make something. Many kinds of services are represented, with health care in the lead, followed by education, and grocery, retail, financial services, and fast food/convenience stores.

The contrast to the intermediate past is shocking. Courtesy of researchers at the Pennsylvania Department of Labor and Industry, here are the top 25 employers of 1965 (the earliest year for which they have available records):

1) United States Steel

2) Bethlehem Steel

3) Westinghouse Electric

4) Bell Telephone of Pennsylvania

5) Jones & Laughlin Steel

6) General Electric

7) Sears, Roebuck

8) A&P

9) Acme Markets

10) Western Electric

11) Philco

12) Budd

13) Philadelphia Electric

14) Boeing

15) Crucible Steel

16) Pittsburgh Plate Glass

17) Allegheny Ludlum Steel

18) Sylvania Electric

19) Sun Oil

20) Pittsburgh Steel

21) Armco Steel

22) Aluminum Company of America (Alcoa)

23) RCA

24) Armstrong Cork

25) Rohm and Haas

(Methodology was not made clear: government entities, hospitals, and universities are not listed, but the absence is unexplained.)

Of the 25, services are only represented by retailers and utilities: no banks or health care providers make the top 40. Seven steelmakers dominate the list, joined by Alcoa. The state's heritage in energy was still represented by Atlantic Refining in Philadelphia, Sun Oil, and Gulf Oil. Transportation is more of a factor today, with UPS at 9 and US Airways at 30; in 1965, no railroads made the list, even though their suppliers (Budd, GE, Westinghouse Air Brake) did.

The composition of the 1965 and 2008 lists illustrate several germane points with regard to the current downturn. First, it's hard to stay on top: almost all companies that at one time appeared to be powerfully untouchable sooner or later fall by the wayside. Bell of Pennsylvania, the #4 employer, morphed into Verizon, presently ranked 28th. Among retailers, A&P disappeared, as did Gimbels and G.C. Murphy, while Sears fell from seventh to 32nd. Second, the stimulus packages of the 1930s -- when manufacturing, agriculture, transportation, labor unions, and foreign trade were unrecognizable even from the vantage point of the mid-1960s -- would appear to present few lessons for current policy-makers. Finally, the kinds of firms traditionally targeted by economic development agencies -- addressing dynamic markets, paying high wages, and anchoring a community or region -- are in the Pennsylvania case not particularly large employers:

14) Merck

19) The Vanguard Group (headquartered outside Philadelphia)

25) Comcast (headquartered in downtown Philadelphia)

39) General Electric (which makes railroad locomotives in Erie).

Are We All in the Same Boat?

Given the speed at which the U.S. economy fell into recession (recall that oil prices dropped $100 a barrel in a quarter), the dust is clearly not fully settled. At this juncture, are there some states that might start attracting internal migration given relatively healthier economies?

According to the Bureau of Labor Statistics, December 2008 U.S. unemployment ranged from a low of 3.4% in Wyoming to a high of 10.6% in Michigan. Given that Wyoming is attractive for retirees, miners, and ranchers, it is unlikely to be an employment destination of choice for the kinds of people being displaced elsewhere; North and South Dakota are in a similar situation. Unemployment remains in the 4% range in Iowa, Nebraska, New Hampshire, New Mexico, and Oklahoma -- tellingly, three of those states formed the heart of the Depression-era "Dust Bowl" memorialized in John Steinbeck's The Grapes of Wrath as a place to leave. Unlike the "new South" in the 1970s and '80s, or Arizona and California over the past half-century, these states a) will not serve as a magnet for the unemployed and b) are sufficiently distinctive that they don't serve as a model for other states to emulate.

Those BLS numbers tell a similarly fascinating story: while Rhode Island's December unemployment (10%) is the worst since statistics were standardized in 1976, nine states recorded their record _low_ unemployment month in calendar 2007-2008, including West Virginia, which had its best unemployment month in over 30 years just last August. All told, the unemployment figures have a long way to rise, even in Michigan, which reported 16.9% unemployment less than 10 years ago, in March 2000.

So who's the exemplar, the state others want to emulate on the jobs front? Considerable debate in the literature revolves around a concept associated with Harvard's Michael Porter, that of economic clusters. As the classic formulation puts it, clusters are "Geographic concentrations of interconnected companies, specialized suppliers, service providers, firms in related industries, and associated institutions (for example, universities, standards agencies, and trade associations) in particular fields that compete but also co-operate." Classic examples can be found in and around Detroit, Palo Alto, Los Angeles (aerospace and Hollywood), and New York/Greenwich (hedge funds). Does cluster theory explain the health of the currently healthiest states?

Not so much. At 6% unemployment in December, Texas falls pretty much in the middle of the pack. Its worst month since 1976, however, is only 9.3%, relatively low and surprising in that Texas has been hit hard by economic crises related to real estate (the savings & loan scandal), low oil prices, Enron, and the bursting of the tech bubble in 2001. Texas combines mineral wealth, oilfield-related services, high tech, agriculture, less location-specific service industries (including American Airlines), and tourism to construct a reasonably well-hedged, recession-resistant economy.

With the exception of California, no other state has so successfully balanced natural endowment, knowledge-intensive industries (including military bases and hospitals), ranching and farming (including some ranches spectacularly successful in attracting agricultural subsidies), and cultural distinctiveness: even the state's anti-littering campaign, admonishing "Don't Mess with Texas," was a huge success, one widely emulated elsewhere. Given all these factors, it would not be surprising to see the state attract some decree of internal migration this time around.

Morals of the Story

-Like politics (and closely related thereto), all unemployment is local. Congress could face massive pressure as the mid-term elections are only 21 months away and the economy cannot bounce back that fast everywhere.

-At the state level of job generation, it helps (as in Virginia) to host lots of federal government jobs.

-Exception to the rule above: when a military base closes, the hurt is broad and immediate. Maine's Brunswick Naval Air Station will lose about 5,000 people (including dependents) as it shuts down, with an estimated annual impact of $200 million.

-For all the appeal of brand, brainpower, and other intangible assets in the "new economy" lore, the healthiest states in the Union right now are sitting on mineral or dirt-based wealth for a portion of a balanced economy.

-Cluster theory has failed to be proven as most university-based economic development efforts relate more to quality of life than direct linkage to a given university's academic focus areas. Stanford provides a glaring exception that proves the rule.

-Cluster theory also generates economic monocultures, which tend more to boom-and-bust cycles than do more diversified economic ecosystems. Clusters are also proving difficult to undo, as Michigan's limited success in luring materials science, biotech, software, and other non-automotive firms illustrates.

The rapid shift of the U.S. economy to a services-driven structure with a massive trade imbalance presents the Obama recovery team with many new challenges. Government employment is already high, and will be limited by falling tax revenues. The M&A activity of the previous decade has generated some very large organizations that, along with Detroit's Big Three, are shedding jobs at a rapid rate. Even though a "knowledge economy" sounds intuitively appealing, at some point U.S. manufacturing will need to be redefined for employment and trade to behave more sustainably. Finally, the biggest intangible of all -- consumer and investor confidence -- will play a critical role in a recovery, and the relationship between $819 or however many billion dollars and that elusive quantity remains to be determined.