Minimum Wage & Unemployment: Confusing Micro and Macro

The recent movement to raise the minimum wage to $15.00/hr. has brought out the usual dire warnings that this will cause a significant increase in unemployment and lock out low wage workers from obtaining jobs. Intuitively, this seems to make sense. When one looks at this scenario from the viewpoint of a business owner, the common sense outcome is you would have to offset the increase in costs by reducing staff.

Classic demand and supply analysis of the labor market from Econ 101 would seem to confirm this as you can see below:

Wages set above equilibrium creates surplus of labor - unemployment.
Wages set above equilibrium creates surplus of labor – unemployment. Image: Wiki Commons.

Yet, the evidence is clear that unemployment does not rise with an increase of the minimum wage. Why should this produce such a counter-intuitive result? The key to the answer lies in the fundamental differences between microeconomics (study of individuals and firms) and macroeconomics (study of the economy as a whole) as well as a more complex model of labor markets developed the past few decades referred to as efficiency wage theory.

First, lets take a look at that classic Econ 101 model of labor markets since this is how the issue is most often debated in the popular media and general public.

Models of micro units in the economy are open systems. Take an employer for example, income flows into the employer from an outside entity (customers). Likewise, spending flows out to entities beyond the employer in the form of wages.  The argument against raising the minimum wage sees the cash flow out increasing without an increase in the cash flow in.  Hence, staff is reduced to offset this outflow.

Employer is an open system. the system is "permeable" as cash flows in and out of the system boundary.
Employer is an open system. the system is “permeable” as cash flows in and out of the system boundary.

The same does not hold for a national economy as a whole. Why? A macro unit, such as nation, is a closed system. As Paul Krugman says, in this scenario, everybody’s spending is someone else’s income. If spending drops overall in a macro unit, then income must necessarily drop as well.  This is the cause of business cycles.  An example of a closed system is below. The three major components of GDP are consumption, investment, and government spending. Unlike a household or business, there is no income to flow in from outside the system.  As we’ll see, modulating these business cycles is more important than minimum wage laws in reducing unemployment for low wage workers.

Closed System
Boundary around a closed system is “impermeable”. Cash flows remain inside the system and do not leak out.

As noted earlier, the classic micro model would indicate that an increase in the minimum wage forces employers to reduce staff and also increases the available pool of labor as higher wages induce more people to look for a job.  In this model, an increase in the minimum wage can represent a transfer of wealth from employers to employees, which is the real cause of the political friction on this issue.  Framed this way, it directly pits employees against employers.

Is this transfer of wealth fair?

In the late 1800’s, Alfred Marshall made the great defense for capitalism against the growing socialist movement. Marshall postulated that increased worker productivity would result in increased wages and that was the key to reducing the great poverty of the time. How to increase productivity? Marshall proposed an expansion of expenditures in public education. He was also recognized the productivity gains acquired from spillover knowledge. That is, less experienced workers increase productivity by working with more experienced workers.

The classical economic model derived by Marshall (and others) suggested that workers wages are commensurate with productivity in a free labor market. This model makes a few assumptions. Among them being:

Information is symmetric. That is, both employers and employees have the same knowledge of the existing labor market.

The labor market is competitive to the point where neither an individual employer nor employee can affect the wage rate.

The economy is always at full capacity.

To paraphrase Harry Callahan, a good economic model has got to know its limitations.

How close to reality are these assumptions? A good diagnostic is to compare productivity gains with labor costs (wages and benefits). If the model is correct, both these variables should match. What does the data show?

Below is a comparison between annual increases in worker productivity and real hourly compensation (which accounts for both wages and benefits) since 1979:

Data Source: St Louis Federal Reserve
Data Source: St Louis Federal Reserve

For the most part, compensation lags behind productivity.  Only periodically has compensation matched productivity and that includes the mid-1980’s and late 1990’s.  Hence, the economy usually operates at less than full capacity.  Since the late 1990’s, compensation has seriously lagged behind productivity.  This represents a transfer of wealth from labor to employers not predicted by classical micro models of the economy.  Consequently, the defense of free labor markets as a means to reduce poverty breaks down.  How to change that?

It helps to use real world case studies rather than fictional work.

One thing to avoid is fear of a supply shock by employers “going Galt” and reducing and/or quitting their businesses in a snit over increasing wages.  During the period between 1947-79, while unions were at their peak, wages kept up with productivity gains.  The result was an expanding middle class and business did just fine.  Employers might resent an increase in the minimum wage, but on an aggregate scale, will not have an incentive to downscale their business unless wage increases surpass productivity increases for a sustained period of time.  If that did not happen in the post-World War II period, it’s not likely to happen now.

One of the major critiques of the minimum wage law is that it locks entry level workers out of the job market by keeping wages above the market equilibrium level.  However, the main driver of unemployment for teenagers (by definition entry level employees), as with any section of the population, is the business cycle seen below:


The impact of minimum wage increases is dwarfed by the effect of the business cycle on unemployment.

Here is where macroeconomics comes into the picture.  Over the past 35 years, teenage unemployment topped 20% three times, all during recessions.  The Great Recession, created by the 2008 financial crisis, produced a teenage unemployment rate of over 25%.  The first step in creating job opportunities is to modulate the business cycle in a manner to avoid steep recessions.  A combination of New Deal banking regulations and appropriate monetary/fiscal policy was successful in this regard from 1947-1972.

It doesn’t make sense to oppose minimum wage laws as a means to decrease teenage unemployment if one is also opposed to employing monetary and fiscal policy to moderate business cycles.  The first priority in this direction is to regulate the financial sector so the risk of banking crisis are reduced.  It is financial crisis that cause periods of severe unemployment lasting 3-5 years, sometimes longer.  Prior to World War II, financial panics induced multi-year depressions in 1857, 1873, 1893, and 1929.  The last recession is not an isolated event, but a natural consequence of an unregulated financial sector.  Younger workers are significantly at risk during these events of long-term unemployment.

An additional step is to index the minimum wage to the inflation rate.  The minimum wage topped out at $10.69/hr (2013 dollars) in 1968 and has steadily eroded since.  Overall unemployment was 3.6% during that year with a teenage unemployment rate of 11-12%, a further indication that the business cycle is a greater determinant of teenage unemployment than the minimum wage.   Also, indexing minimum wage to productivity increases needs to be considered.  Paying for production is a reasonable proposition.

The perfect labor market model as presented by the demand and supply graph at the top of this post is an abstract concept.  That model relies on assumptions that cannot be fully realized in the real-world.  Think of it as the economic version of the Carnot engine, which represents the theoretical limit for engine efficiency.  You cannot build such an engine in the real world as it relies on a cycle that does not lose heat to friction.  Likewise, it is impossible to build a perfect labor market in the real world as efficiency is lost due to frictions such as incomplete information and an economy operating at less than full employment.

For an economist to claim free labor markets are efficient to the point where labor receives rising compensation with rising productivity is the same as an engineer who claims to have built a 100% efficient engine.

We need to realize labor is not the equivalent of widgets.  Classic demand and supply curves oversimplifies human behavior in the labor market.

However, a new, more complex theory of labor markets has emerged over the past few decades that merits real consideration as it predictions coincide with some real world observations.  And this is efficiency wage theory.  This theory predicts that employers have a menu of wages to pick from rather than a single market wage.  The tradeoffs involve low wage, low productivity, high employee turnover, vs. high wage, higher productivity, and low turnover.

Lets take a look how employers have responded to an increase in the minimum wage. For starters, layoffs are not the primary, or even a significant reaction. A variety of strategies are employed to offset the higher cost of labor. One is to train employees in a manner to boost productivity. Another is to reduce costly employee turnover. Also, some employers elect to reduce profits to cover the cost.  The real response to minimum wage increases are more reflective of the efficiency wage concept than the classic single wage demand and supply model.

New entrants in the labor market need to acquire social connections to move into higher wage brackets, even within the same skill level jobs.

A 1984 survey paper by current Fed chair Janet Yellen noted that efficiency wage theory predicts a two-tiered workforce.  One is a high wage workforce where jobs are obtained mostly via personal contacts.  Employers have a comfort level with employees they personally know and do not feel the need to go through expensive vetting and monitoring processes.  The other tier is a low wage, highly monitored workforce with a lot of turnover.  Sound familiar?  The latter seems to reflect low paid contract/temp workers who often perform the same functions as higher paid permanent employees at the same company.  Here again, the efficiency wage model seems to trump regular demand and supply.

It does appear to be time for policy makers to incorporate this more sophisticated model when addressing low wage workers and the chronically high unemployment rate in various demographic groups of the workforce.  In particular, is the need to promote a way for low wage workers to make the leap into the high wage sector.  As noted in Yellen’s paper, ability and education is not enough, social connections play a key role in obtaining a high wage job.  This, combined with proper fiscal/monetary policy, brings the best hope for lifting individuals from poverty.

It is often said that anyone who has taken Economics 101 will understand raising the minimum wage causes unemployment to increase.  The efficiency wage model is a bit beyond Econ 101, probably at an intermediate level course.  And that’s perfectly fine.  If you are diagnosed with a serious illness, do you want to be treated by a doctor who went to med school or is using Bio 101?  The same holds true in economics.

*Image on top is the Chicago Memorial Day Massacre of 1937.  Ten striking workers were killed and gave momentum for the Fair Labor Standards Act of 1938 which mandated the first minimum wage.  Photo:  U.S. National Archives and Records Administration.

Leave a Reply

Your email address will not be published. Required fields are marked *