It’s the econom … uh, state, stupid!

by Jehu

So, let’s put together the arguments from the previous two posts on hours of labor and the minimum wage.


Here is the minimum wage from 1938-2016 in nominal terms

CHART: Change in the official unadjusted hourly minimum wage 1938 to 2012

And here is minimum wage (1939-2016) adjusted for its purchasing power in terms of gold

CHART: Change in the official hourly minimum wage 1938 to 2012 adjusted for the purchasing power of dollars as measured by gold

As you can see, according to the second chart the purchasing power of the minimum wage (as measured by gold) peaked in 1970 and fell sharply after — never recovering its 1970 purchasing power. Based on gold, all subsequent increases in the minimum wage failed to keep up with the declining purchasing power of the currency.

Assuming wages influence employment, we would expect to see aggregate working time to reflect the declining purchasing power of the currency. This is just the corollary of the proposition higher wages reduce aggregate employment. If higher wages reduce employment, and if the relation is commutative, we should expect lower wages to increase aggregate employment.

And this is exactly the argument Benjamin M. Friedman makes in his paper, “Work and consumption in an era of unbalanced technological advance”.

“Until the 1970s, Keynes was right on both fronts: per capita output grew at the upper end of the range he predicted, most families’ incomes grew even faster (inequality was mostly narrowing during that period), and the workweek continued to decline. But with widening inequality from the early 1970s on, the growth of most families’ incomes became far slower than he had predicted, and the workweek stopped declining. The latter combination has persisted ever since.”

According to Friedman, hours of labor stopped declining at exactly the point where the minimum wage, measured in gold, collapsed. Friedman’s argument is likely a scientifically valid hypothesis only if the theory that higher wages reduce employment is both true and commutative.

So does this data amount to empirical confirmation of the theory higher wages reduce employment? And, if so, what are the implications of this theory?

In 2014, Lauren Carroll tested the hypothesis raising the minimum wage increases aggregate employment for Politifact. She defined the hypothesis that an increased minimum wage increases employment as true if, since 1978, aggregate employment increased in the 12 months following a legal increase.

There were eleven increases in the minimum wage between 1978 and 2016, raising it in aggregate from $2.30 to $7.25.

Six data points showed aggregate employment increased, while five showed aggregate employment decreased. According to Carroll:

“Increasing the federal minimum wage isn’t always followed by national job growth, or loss for that matter. There were 12 solid months of job growth following the 1978 increase, but 12 months of job loss following the 2008 increase.”

However, of the six that showed an increase in employment, three occurred during an expansion; and of the five that showed a decline in employment, three occurred during an economic contraction.

Economists explain this inconclusive data by either asserting an increase in the minimum wage has no impact on aggregate employment, or by suggesting other economic factors swamp whatever impact an increased minimum wage does have.

However, there are reasons to question Carroll’s methodology in making her calculations. First, Carroll does not adjust the minimum wage for any change in the purchasing power of the currency paid out as wages. Based on her data, we have no way of knowing whether any currency increase in the minimum wage actually increased the real purchasing power of the minimum wage.

For instance, as the table shows there is a nine year interval between the increase of 1981 and 1990, and ten years between the increase of 1997 and 2007. Carroll ignores the impact the appreciation or depreciation of the purchasing power of the currency had on real wages during that time. However, between 1981 and 1990, for instance, the purchasing power of the dollar in terms of gold actually appreciated significantly. And between 1997 and 2007, the dollar severely depreciated, losing more than half its purchasing power against gold.

Carroll knows the purchasing power of the currency changes over time, but she introduces no mechanism to control for this change. Moreover, she does not correct for the actual impact of an increase in the minimum wage on the wages of the working class, which, she admits, appears deliberately designed to effect as few workers as possible.

Carroll has offered up an explanation for why no actual legislated increase in the minimum wage has negatively affect aggregate employment. Moreover, she has ignored the broad sweep of data that, if anything, points to the exact opposite conclusion of the economists she cites. Take this choice piece of disinformation for example:

“One of the most thorough reports on this debate comes out of the Center for Economic and Policy Research, a progressive think tank. The report concludes that the “bulk of the best statistical evidence” shows little to no effect on employment when the minimum wage goes up.”

What evidence does she offer that the real minimum wage has ever actually gone up? In fact, measured in gold, the minimum wage has been declining since 1970 and even by official government measures of inflation has a purchasing power that is almost 50% below where is was in 1970.

Carroll can’t even establish her assertion the minimum wage has increased, much less what, if any, impact it has had on aggregate employment.

We are thus offered two theories of the relation between wages and aggregate employment from two different sources focusing on two different presentations of the same problem. In the first, offered by Carroll, wages have, at best, an ambiguous relation to aggregate employment. In the second, offered by Friedman, declining wages account for why labor hours stopped declining in the 1970s. In the first theory, the change in wages has no impact on aggregate employment; in the second, aggregate employment is explained by wages.

You can pick and choose which ever theory best fits your argument. If you want to advocate for a higher minimum wage, you can choose Carroll’s explanation, where aggregate employment is unaffected by wages. If you want to argue against less labor, you can borrow from Friedman’s explanation, where poverty makes less labor impossible.

What both arguments overlook is the role the fascist state has played in the relation between wages and aggregate employment. Neither the legally mandated minimum wage nor aggregate employment are generic transhistorical categories; they are political categories. The state defines the legal minimum wage and it defines the aggregate employment level. Yet, state policies remain the unmentioned factor in the discussion of the relation between wages and employment.

To put this another way, Ben Friedman asks an important question: Why was Keynes wrong in his prediction of a 15 hour work week by 2030? To answer this question he mines the empirical data showing a marked decline in wage income beginning with the 1970s. Friedman is convinced the answer to his question is to be found in the empirical data, if only he can locate it amidst the morass of official government statistics.

In fact, as a Harvard economist and member of the National Bureau of Economic Research, he could have just asked former president of Harvard and fellow Harvard economist and fellow NBER heavy, Larry Summers, the question. Had he done so, Summers would have likely offered Friedman this answer:

“The primary objective of our policy is having more work done, more product produced and more people earning more income. It may be desirable to have a given amount of work shared among more people. But that’s not as desirable as expanding the total amount of work.”

According to Summers, then, it is the policy of the United States government to expand the total amount of labor we perform. This raises the significant likelihood the real minimum wage has been deliberately set to level consistent with the policy goal of maximizing aggregate employment.

Now, I suppose it is possible Friedman and Summers have never met despite both being at Harvard and members of the NBER. They never passed in the hallways or attended the same functions or could even recognize each other across the room. But I do assume both of these simpletons can read and Summer’s argument is in black and white on the pages of the Washington Post.

Since I am clearly an idiot, I am forced to note that historically almost every general reduction in hours of labor has been legally mandated by government. If I were looking for the reason why hours of labor have not fallen in almost 50 years, I would naturally begin by asking why government didn’t reduce them. I certainly would not begin by asking why my neighbor did not reduce her own hours, but why government did not reduce everyone’s hours.

But, again, I am an idiot that way. I think the answer to why hours of labor haven’t been reduced lies with government, not my neighbor. Since government legally mandates hours of labor and since it legally mandates the minimum wage, I think this is only right.