Note: See also
- The State of Colorado Employment, showing graphs of "actual" vs. "official" unemployment.
- Fed Policy, NAIRU, and the "Phillips Curve" for papers that explain how Federal Reserve policies are biased against those who work for a wage and describe the flaws in the "Phillips Curve" theory and the concept of NAIRU (Non Accelerating Inflation Rate of Unemployment). As long as full employment is below NAIRU (on the order of 5 or 6%, which the Fed determines), the Fed considers we have "full employment", which does not mean zero unemployment. While the purported purpose of a 5 or 6% NAIRU is to allow for job "churn", the purpose is actually to favor return from capital over return for labor.
- Sections of Response to a Conservative that more fully document Fed policy and its effect.

"Free market" ideologues believe that wages are determined by labor market supply and demand. It isn't true, but it's such a powerful belief that they ignore, or don't even recognize, national policies that subvert this simplistic notion.

This article explains why this free market fantasy is false. It begins by examining growth and infrastructure backlog issues and then examines unemployment. They are related, so please be patient.

The "free market for labor" myth

The "free market for labor" myth is that unions and a minimum wage are interference in the free market. While this is superficially true, these are actually responses to the already prior and overriding interference in the economy by the Federal Reserve. The interference? Federal Reserve policy rigs demand to assure there are always more people who want jobs than there are jobs.

This understanding of how Federal Reserve Board policies affect employment and unemployment came to me as I worked to understand the dynamics of growth. I've come to realize that over the long run:

• regions have infrastructure backlogs for the same reason that so many people have low wages
• real unemployment is much greater than official statistics lead us to believe ... for more on this see Unemployment: Official, Effective, Real, Jobs & 'Trade' Data Update Apr08, and Shadow Government Statistics at John Williams' Employment and Unemployment Reporting link.
• Federal Reserve policies drive wages at the bottom to somewhere between zero and subsistence level. That's bcause significanlty more demand for jobs than jobs produces a game of "musical chairs." This is described in the book, Co-opetition by Brandenburger and Nalebuff, on applications of game theory to business strategy in the section on Game Theory, pp. 40 - 44. In the following pages they describe how to change the game: negotiate as a group ... "change the game by banding together" (p. 47).
• making up for the inequities caused by Fed policies is the reason there's a minimum wage
• for the labor market this means unions are the only way to overcome the Fed's fixed game and stand a chance against "unions of capital" (i.e., corporations).

The most common tactic of what’s called "economic development" is to lower taxes and increase subsidies to businesses to attract them. But the result is an escalating "my region can impose lower taxes and regulations than your region" competition between regions.

Because Fed policy allows only so many jobs to be created, it's a zero-sum game for states to compete by giving incentives to companies to get them to locate in their state. Art Rolnick, an economist at the Federal Reserve Bank of Minneapolis, also points out that Congress Should End the Economic War Among the States because it results in a negative return on investment.  His recommendation instead? Invest in Early Childhood Development on a Large Scale for a positive return on investment.

The result of these "tax wars" is a nationwide infrastructure backlog of $1.6 trillion. this backlog is rising at a rate of 9.25% per year, many times the rate of inflation.

And, as we can all see, it’s happening in Colorado Springs, too. In Nov. 2000 Dave Zelenok, Group Support Manager Public Works, said that, "if things keep going the way they are, we’ll be facing a $3 - 4 billion backlog" in Colorado Springs in the next 10 - 20 years.

Now it’s logical for every region to want to compete by lowering taxes and regulations … it’s called being "business-friendly." But a "logical for every individual region" strategy leads to infrastructure backlogs and declines in efficiency and competitiveness for every region and therefore for the nation as a whole.

Logical actions in each individual region are collectively insane. This is an example of what economists call the "bounded rationality" of human decision-making.

As regional infrastructure backlogs grow and services decline, regions are led to promote more development in order to gain tax revenue. This provides some immediate relief, but years later the increased load on the infrastructure creates an even larger backlog. This prompts regions to promote even more growth for more immediate relief.

This is called "addiction." It is the same structure as addiction to drugs: feel good in the short term even though long term health suffers. Because there’s a long delay before the infrastructure demands arise, it’s easy to ignore, and/or obfuscate, the connection.

Of course, addiction of any kind is not a solution. No one can sell a product at a loss and make it up in volume.

Fine, but what's this got to do with jobs and unemployment?

Regions are virtually forced to compete because U.S. government policy only allows so many jobs to be created in the United States.

In its drive to control inflation, the Federal Reserve Board keys monetary policy and interest rates to a NAIRU (Non-Accelerating Inflation Rate of Unemployment), which the Fed generally believes is on the order of 5 or 6%.

For example, if the Federal Reserve fears inflation from any cause and believes too many jobs are created, or believes unemployment is too low (below its NAIRU target), the Fed raises interest rates or shrinks the money supply to slow the economy and reduce demand for labor. There’s lower demand because there’s less investment and fewer people working and therefore less upward pressure on prices.

It used to be that the Fed thought NAIRU was 6% or more. But in the 90s the graphs below show Fed policies produced substantial job growth and allowed unemployment to go down to 4% thanks to fiscally responsible tax increases in 1993. While wages went up slightly, this confirmed the vast reservoir of labor available to meet the greater demand.

There was little wage inflation when unemployment was down to 4% during the Clinton years. Which economy would you prefer?

A higher rate of job growth drove unemployment down during the Clinton years.

[Note: BLS data in Apr 2005 shows 275K more jobs added between Jan 2000 and Aug 2004 than were shown in BLS data from Sep 2004. Strange or Fudged? See below the BLS response to my inquiry as to why.]

So employment over the long run depends on the NAIRU (in the short run it depends on aggregate demand). The Federal Reserve pursues this policy in the belief that it must do so to avoid an inflationary wage-price spiral (increasing prices resulting in higher wages and even higher prices, etc.).

Some don’t believe the Fed reacts to employment in this way, but the stock market "knows." When the economy is strong, investors tend to sell when there is either a good unemployment report (unemployment falls) or a good "jobs report" (an expansion of the number of people employed), because they know the Federal Reserve is likely to raise interest rates to "cool the economy."

Fed interest rate increases burst the stock market bubble and also hurt the economy (instead of raising interest rates, the Fed should have increased margin requirements to stop speculation).

Some also don’t believe the Federal Reserve has enough power to affect the economy this way. But the stock market does respond to Fed power. As an example, the chart shows the Fed funds rate and stock prices. Raising interest rates quite nicely burst the stock market bubble of the 90s.

The total number of jobs in the U.S. is determined by Fed policy, with the result that nationally there are always more people than there are jobs. So tax competition between regions does not "create jobs;" it simply shifts jobs among regions. It creates higher growth in some regions and lower growth in others, but it does not decrease (overall) unemployment.

I didn’t believe this until I plotted the graph below showing that regions with higher growth do not on the average have greater reductions in unemployment.

It shows that higher growth rates in Metropolitan Statistical Areas between 1997 and 1998 did not produce a more positive "change in employment" between 1998 and 1999. The reason is that people who are out of work in one region move to higher-growth regions.

Regions with higher growth rates don't show a reduction in unemployment..

Find more graphs of "unemployment" and "change in unemployment" vs. regional growth rates at the end of this file (pdf, 178K). There's much scatter in many of the graphs and no clear trend. Many "unemployment" graphs do show reduced unemployment, but continued low unemployment would require unsustainable growth rates.

Real unemployment

This effect might seem more dramatic than one would expect for a 5 or 6% official unemployment. However, official unemployment is vastly understated.

As shown in the figures below, because job growth hasn't kept up with population growth and because some categories of unemployed aren't officially counted, unemployment is more like 10%.

This shows that US employment is not keeping up with population growth. There's a 4.5M backlog.

And as the figure below shows, "official" unemployment, U-3, is a vast understatement.

This shows that US unemployment is much, much greater than the "official", U-3, statistic ... and this doesn't count the underemployed.

A graphic illustration that there is vast unemployment, understated even by these graphs: 25 thousand apply for 325 jobs at a new Wal-Mart in January 2006 (another link to this article or google: "325 jobs" Wal-Mart). Another graphic example: In March 2007 in Detroit 26,000 apply for 1,000 casino jobs.

Unemployment is typically understood to be higher in Europe than in Japan and the U.S., but data shows that rates of unemployment are not all that much different ....

                                         U.S.    Japan    Britain    France
Unemployment at the
end of 1993:       Official:     6.4%     2.9%    10.2%    12.0%
                       "Effective":     9.3%     9.6%    12.8%    13.7%
     effective/official ratio:     1.453     3.310     1.255     1.142

Official unemployment only counts those who are actively seeking work.  "Effective" unemployment includes "discouraged" & "involuntary part-time" workers (but not under-employed).  Britain and France are much more efficient in counting all unemployment.  [Reference: Amex Bank Review analysis using BLS data and cited in The Economist, 2/5/94, p. 25]

And this doesn’t even include other categories that provide sources of labor. Lester Thurow in an interview on his 1996 book, The Future of Capitalism, estimated slack in the labor force at more like 30% (including those officially counted, 5.7%, don’t meet the official test, 4.5%, part-time, 3.4%, on-call, 1.5%, disappeared, 4.5%, and self-employed — many of whom are underemployed, 6.1%).

And we can’t compare a 6% unemployment today to a 6% unemployment in the 80s because contingency workers — part-timers, temporaries and contract workers who lack full benefits and job security — make up a greater part of the labor force, accounting for as many as 30 million of the workers (out of ~137 million employed). The number of mainly low-wage temporaries has tripled over the last dozen years (Denver Post, 9/5/94).

It shouldn't be surprising that the "official unemployment" data is an underestimate. No government would want to admit that national policy is set to assure that on the order of 20% of the workforce (1 in 5) is unemployed or underemployed.

Because 20 - 30% of the working population wants work or more work or work for which they're better qualified, regions are essentially playing "musical chairs." Because some regions are sure to not have enough jobs, the added value of any one region zero.

So because there is only so many jobs and only so much economic growth allowed, companies can say, "Give us a deal, or we’ll go elsewhere." Companies can demand concessions, just as a sports team demands that a city build a tax-payer-funded stadium, if it wants the team to move there. In the book, Co-opetition, by Brandenberger and Nalebuff, they call this "sacking the cities." The counterpart for Fed policy is "sacking the workers."

So when jobs shift from one area of the country to another driven by "economic development" tax wars and subsidies, there's churning as people move across the country to follow those jobs, but no more total jobs are created.

The Fed assures there's a zero-sum game competition among regions that hurts all regions.

Because there are more people than there are jobs, the added value of any one person is also zero.

Employers can say, "Some people are going to be without a job, so take the job at this wage or someone else will." This is why wages are stagnant at the bottom and why there's a minimum wage. Many say that a minimum wage is an interference in the "free market" for labor, but they ignore the Fed’s prior interference. There is no free market for labor.

Forming a union and bargaining together is the only way for workers to overcome this "musical chairs" dynamic in which some will always be without jobs and the added value of any one worker is zero. Unions prevent companies from playing one worker off against another using the argument: "Someone's going to be without a job ... do you want it to be you? ... take it at this wage or else it will be you."

It's this dynamic that's led to increasing profits, but falling wages:

In the past, higher productivity has translated into higher wages and more jobs as employers share the gains with workers. But this time that hasn't happened yet. Instead, the returns from higher productivity are going into higher profits and lower prices. Using official Bureau of Labor statistics, Johns Hopkins University economist Arnold Packer calculates that employees' share of the value added in the U.S. economy has fallen to its lowest point since records were first kept in 1947 -- and the rate of decline is accelerating.
“Maybe We Could All Deliver Pizza . . .,” Jodie T. Allen, 3/06/04, The Washington Post, http://www.washingtonpost.com/wp-dyn/articles/A35469-2004Mar6.html

In addition to Fed policies, labor laws have also undermined union strength. So-called "right to work" laws were passed to undermine union strength by letting some employees "free ride" on those who do pay dues; the equivalent in our democracy would be letting people opt out of paying taxes and still get the benefits of roads and national defense.

Furthermore, capital organizes itself in corporations and a labor union organization counterpart is a necessary counterbalance to corporate power. It's difficult to understand the antipathy to unions because corporations are virtual dictatorships, whereas unions are democracies. So unions have challenges that corporations do not; democracies are much more difficult to manage.

It really doesn't make sense that "union bosses" are so villainized. After all "corporate bosses" have more power than "union bosses;" look at how CEO pay has increased, even for companies that subsequently failed. .

So there’s an important analogy between taxes and wages. Taxes can be considered "regional wages" that allow regions to maintain a certain quality of life, just as wages allow people to maintain a certain quality of life. Regions have infrastructure backlogs for the same reason that many people do not make a living wage.

This is a surprising example of how we’re all in this together. People stuck in traffic, people in low wage jobs, and tax limitation advocates all have complaints that originate from the same structural cause: Federal Reserve policy that suppresses demand for workers.

But these groups are usually in conflict because they don’t understand this. Think what would happen if they joined forces to push for policy changes that would actually solve their problems!

For more on growth, see the Growth Facts of Life and the much more detailed Tangle of Growth.

Addenda:

1. In the 49 months since the recession began in March 2001 through May 2005, there has been a net increase of 782,000 jobs. To keep up with population growth over this period, 137,000 jobs must be added every month. Over 49 months, that's 6,713,000 jobs that are needed just to stay even. That means that over this period there's a job backlog of 5,931,000 jobs. That 6 million job shortfall never seems to get attention in the media.

2. Somehow the official BLS data in May 2005 shows there were 275,000 more jobs added from Jan 2000 through Aug 2004 than was the case for the same data series that was official in Sep 2004. [Go to http://data.bls.gov/cgi-bin/surveymost?bls and retrieve Series CES0000000001 for the current data. Had I not saved it, I wouldn't have the data as presented in Sep. 04.]

Where did the extra 275,000 jobs came from? The response from BLS, for which even more research would be required for a clear explanation:

From: cesinfo <cesinfo@BLS.GOV>
To: "'scuba@usa.net'" <scuba@usa.net>
Cc: cesinfo <cesinfo@BLS.GOV>, "Duffin, Joshua - BLS" <Duffin.Joshua@bls.gov>
Subject: CES Response: Job Numbers
Date: Wed, 1 Jun 2005 09:14:28 -0400

Dear Mr. Powell,

The revisions you describe are the result of our annual benchmarking process.  Every year, with the release of employment estimates for January (in the first week of February), we revise the previous year's estimates to reflect more current universe counts of employment (mainly based on data from Unemployment Insurance tax reports from employers), and revise up to the previous five years of estimates with newly calculated seasonal adjustment factors incorporating the most recent data available.

More information on the benchmark process is available on our website at http://stats.bls.gov/ces/home.htm#benchmark, including the detailed article at  http://stats.bls.gov/web/cesbmart.htm, and the description of the most recent benchmark revision at http://stats.bls.gov/ces/cesbmk05.htm.

Please let us know if you have further questions.

Sincerely,
Joshua Duffin, Economist
Division of Current Employment Statistics
Bureau of Labor Statistics
duffin.joshua@bls.gov
202-691-6555