Why Economics Failed: The Parable of the Minimum Wage

economist mod econ theory

(Illustration by John Berkerly for The Economist, July 16, 2009)

“Macroeconomics” is, of course, to be distinguished from “microeconomics.” Macro-economics deals with economic affairs “in the large.” It concerns the overall dimensions of economic life. It looks at the total size and shape and functioning of the “elephant” of economic experience, rather than the working or articulation or dimensions of the individual parts. To alter the metaphor, it studies the character of the forest, independently of the trees that compose it.  — Gardner Ackley, Macroeconomic Theory,  Macmillon, 1961, 1963,  (p. 4)

An eminent member of the University of Michigan faculty, Gardner Ackley was an important and influential economist. He was appointed by President Kennedy to the Council of Economic Advisors, and promoted to Chairman of the CEA by Lyndon Johnson. Fifty years ago, however, he may have been even better known in academic circles for his legendary and groundbreaking textbook on macroeconomics. 

It was my intense interest in macroeconomics that propelled me into the PhD program at the University of Michigan following graduation, with honors in economics, from Oberlin College in 1966. At Oberlin, I had learned about macroeconomics from Ackley’s text, and his approach to economics — his incisive logic and unpretentious objectivity — had intensified my interest in the field. Ackley had convinced me that, although there may never have been a “bright line” between microeconomic and macroeconomic theory, there are crucial differences that must be understood and respected between the way an aggregate economy behaves and the behavior of individuals and firms. 

When I was in graduate school at Michigan, Gardner Ackley was away, but I did have one memorable encounter with him, and it was the highlight of my academic experience. In 1970, my senior year of law school, I wrote a paper for a labor seminar discussing the incompatibility of full employment and price stability — stagflation. The paper is gone now, and I don’t remember exactly what I wrote, but I do remember emphasizing structural unemployment and the impacts of the Vietnam War; and I remember that it earned an “A” from Gardner Ackley, the guest reviewer my professor had asked to read it.

Recently returned from his assignment as Ambassador to Italy, Ackley had completed his years of public service. It did not come to light until many years later that in 1966 he had recommended to President Johnson that he raise taxes to pay for the Vietnam War (here). For the delay until 1968 in passing the recommended tax increase, Paul Samuelson later said, “we paid dearly in the inflation of the 1970s” (here).

Rereading his textbook today reveals that Ackley’s influence on me and the way I understand economics is incalculable. I am forced to wonder whether my own frequent use of the forest metaphor for macroeconomics, a full half-century later, might be traceable to an embedded memory of the quoted passage. In truth, his use of the metaphor was not entirely apt: He not only showed that the line between macroeconomic and microeconomic theory was hazy and occasionally non-existent, but he also adeptly traversed the ground between the “trees” — the mathematical and graphical representations of complex macroeconomic models and theories — and the “forest” — the broad high-altitude perspectives and implications of ground-level analyses. Ackley clarified what economic theory tells us will happen under specific assumptions, and he carefully described the limitations imposed on the use of a model by its underlying assumptions and by the limitations of data. 

The Descent into Confusion

Today, mainstream macroeconomics is in a meltdown. It has no satisfactory explanation for the steady rise of income and wealth inequality over the last 30-35 years in the United States or the consequences of that long trend, and in the last few years it has been struggling even to make reliable “short-run” forecasts. Long-run growth forecasts remain inherently unreliable, little more than extensions of recent trends in unemployment or GDP. I have for several years been reviewing the history of economic theories to help pinpoint what has gone wrong with economics, and it turns out that Ackley’s insights, fifty years after he published them, can make a huge difference in explaining and justifying a new framework for a “distributional macroeconomics” and reconciling it with Keynesian macroeconomics. 

That progress in “scientific” macroeconomics had already stalled somewhere along the way from Adam Smith to John Maynard Keynes, and never recovered, is now becoming indisputable. Consider the recent frank concessions of Raj Chetty, a Harvard economist who in 2013 was awarded the biennial John Bates Clark medal, an honor bestowed on the the American economist under the age of 40 who has made the most significant contribution to scholarship in the field. In defending the “science” of economics (“Yes, Economics is a Science,” The New York Times, October 20, 2013, here), Chetty conceded that mainstream economics is still struggling to answer the basic questions of macroeconomics, namely, the causes of growth and decline, and that the profession does not yet adequately understand the mechanics of how market economies work:

It is true that the answers to many “big picture” macroeconomic questions – like the causes of recessions or the determinants of growth – remain elusive.

I have been emphasizing this concession ever since (e.g., see my December 2013 post “Economics: The Lost Science,” here), in light of the persistent decline of mainstream, popular economics into ideology:

That this should still be so is amazing, when the basic features of a market economy’s functional mechanisms had occurred, untested, to Adam Smith 237 years ago (Wealth of Nations, Book II, Ch 3). Yet we see today a discipline almost entirely dominated by ideological fantasies like “trickle-down,” an upside-down notion which unabashedly proclaims “less is more” and, somehow, enlists support from the victims of the extreme inequality it engineers.

The continuing confusion surrounding “big picture” issues that I so superciliously dismissed as “amazing” is actually understandable (though not excusable) for several subtle reasons that are revealed in Ackley’s analysis. After rereading his book, I immediately began writing a post for this blog to explain how Ackley’s legacy can help us find the “elusive” answers to Chetty’s “big picture” questions, a post that quickly turned into a two-part series. Two days ago, however, a strange parable on the minimum wage began to develop in my daily routine, suggesting that this post on the minimum wage would serve as an ideal introduction to the other two posts, which will follow soon.

Credit Chetty for distinguishing between “the causes of recession,” which have routinely been addressed since John Maynard Keynes as short-run, cyclical phenomena, and the determinants of long-term growth. Economic phenomena are complex and messy, and any economic model is necessarily an oversimplification. Economic models employ simplifying assumptions to focus on the most important variables, but these assumptions often conflict so much with what actually happens in real life experience as to call into question the validity of the models themselves. That is exactly what was going on when Keynes developed his General Theory of Employment, Interest and Money. Keynes put it this way: 

The celebrated optimism of traditional economic theory, which has led to economists . . . . [teaching] that all is for the best in the best of all possible worlds provided we will let well alone, is also to be traced, I think, to their having neglected to take account of the drag on prosperity which can be exercised by an insufficiency of effective demand. For there would obviously be a natural tendency towards the optimum employment of resources in a Society which was functioning after the manner of the classical postulates. It may well be that the classical theory represents the way in which we should like our Economy to behave. But to assume that it actually does so is to assume our difficulties away. (Ch. 3, “The Principle of Effective Demand”)

Gardner Ackley’s analysis showed how both the classical and Keynesian models depended on demand (spending) to achieve short-run full employment “equilibrium.” His book covered the field as he found it in the early 1960s, and it revealed that the profession’s understanding of long-term growth was still immature and unsettled, and that little was as yet understood about the determinants of overall prosperity and growth. That this was still true in 2013 has been confirmed by Raj Chetty. Indeed, mainstream economics is even incapable today of properly assessing the effects of the minimum wage on employment and growth.

Growth, and the Minimum Wage

My parable begins early Thursday morning, just after 7:00 a.m. I had been up well past midnight working on my discussion of the Ackley legacy, and diminishing returns had set in. I arose at 6:00 however, because I had a 7:00 a.m. appointment for a tooth extraction. When I arrived, my dentist went right to work, giving me a series of shots. He then left me alone for a while to let the numbing process take effect. His assistant handed me the TV controller, and I switched through the channels until landing on C-Span, which was broadcasting “The Washington Journal.”  

Greta Brawner was hosting this broadcast of the The Washington Journal (Thursday, May 1, 2014, here).  Reviewing her program later, I noted that she had headlined the broadcast with a story of declining growth in the first quarter of 2014, in which growth had slowed to an annualized 0.1 %, attributable according to her source to an especially cold winter. While waiting for the anesthetics to kick in, I had switched to C-Span somewhere around 7:15, finding Ms. Brawner’s call-in show in progress, and she was asking viewers for their opinions on how to stimulate economic growth. 

This ought to be good, I thought. I did not yet know her name, but it seemed remarkable to me for the host of a major news program to be asking for views on the causes of growth from the general public, which on average has virtually no knowledge of economics, and no notion whatsoever of the mechanics of growth. It seemed especially ironic to be soliciting such uninformed views at a time when the mainstream economics itself had admitted to finding an understanding of the mechanics of growth “elusive.”  

Ms. Brawner had a stack of news story enlargements in front of her, and when someone called in with a point she wanted to pursue she read quotes from these articles, and people who wanted to emphasize private sector freedom over government intervention, it appeared, were getting more of her time and attention. Before long a caller identifying himself as a political independent offered an unusually comprehensive response. He said he would raise the minimum wage, close tax loopholes for billionaires and corporations, hold corporations responsible for environmental damage, and invest in infrastructure and education.

Well prepared for such a call, Ms. Brawner turned immediately to the issue of the minimum wage, citing a recent Congressional Budget Office (CBO) report concluding that increasing the minimum wage to $10.10 would actually reduce employment by somewhere between zero to one million jobs. Shortly thereafter, the dentist returned and began to extract my tooth.  Later that morning, I retrieved the February, 2014 CBO report “The Effects of a Minimum-Wage Increase on Employment and Family Income” (here), scanned it, and saved it for further review. I replayed the Brawner show, discovering that she had also run a lengthy clip of Mitch McConnell, in an address at the Senate, firmly berating Democrats for hurting the people they claim to be trying to help: We all know, McConnell insisted, that the proposed increase in the minimum wage would cause the loss of one million jobs. 

Why Economics Fails

Fortuitously, the next morning’s New York Times arrived with Paul Krugman’s Op-ed entitled “Why Economics Failed” (here). Krugman had been teaching a class on “The Great Depression: Causes and Consequences.” It was fun, he reported, but:

I found myself turning at the end to an agonizing question: Why, at the moment it was most needed and could have done the most good, did economics fail?

He focused on topics he frequently addresses, including budget deficits, interest rates, and inflation, but his core message was about decline, the opposite of growth:

The financial crisis and the housing bust created an environment in which everyone was trying to spend less, but my spending is your income and your spending is my income, so when everyone tried to cut spending at the same time the result is an overall decline in incomes and a depressed economy. And we know (or should know) that depressed economies behave quite differently from economies that are at or near full employment.  

Coincidentally, Krugman’s explanation of the core problem of insufficient demand was given in terms of the failure of “Say’s Law,” as (I have recently discovered) it had been reframed by Garner Ackley 50 years ago. Supply does not automatically create its own demand, and therefore the failure of economics to which Krugman refers is its failure to recognize Keynes’s “principle of effective demand”: Just as reduced demand causes decline, increasing demand is needed to stimulate growth, as the caller on whom Brawner had unleashed McConnell’s tirade the previous morning appeared to understand. Krugman continued:

And the diagnosis of our troubles as stemming from inadequate demand had clear policy implications: as long as lack of demand was the problem, we would be living in a world in which the usual rules didn’t apply. In particular, this was no time to worry about budget deficits and cut spending, which would only deepen the depression. When John Boehner, then the House minority leader, declared in early 2009 that since American families were having to tighten their belts, the government should tighten its belt, too, people like me cringed; his remarks betrayed his economic ignorance. We needed more government spending, not less, to fill the hole left by inadequate private demand.

Given that government programs consistently redistributing money to people who will quickly spend most or all of it (like unemployment insurance and food stamps) stimulates growth, it would seem to follow that similar private sector wage increases, and a higher minimum wage, would do the same.     

Reflections on the “Law of Effective Demand”

I call Keynes’s principle of effective demand the law of effective demand because it is as close to an inviolable “law” as economics has got. People must have money in order to buy things.

I finished my coffee, put down the New York Times, and headed off to a an appointment I had with my podiatrist that morning. He asked me about my work in economics, which we had discussed on an earlier visit, and I told him that I had learned about the February 2014 CBO conclusion that increasing the minimum wage would substantially reduce employment, and that I needed to study that report closely:

“When you think about it,” he replied, “that makes sense.”

“Why?” I asked.

“If their wage costs are going up, employers will have to cut back on production and jobs in order to stay profitable.”

“Perhaps,” I responded, “but it’s not that simple: Employer responses to short-run cost increases are only a part of the issue. There is also an economy-wide stimulation from all of the additional money circulating in the economy as a result of paying the minimum wage, creating more jobs .”    

His eyebrows rose. He smiled and rubbed his chin, and told me he wanted to see what I was writing about economics.   

The CBO Report

Pages from CBO effect of Minimum Wage on employment feb 2014

In its first figure (p. 5) CBO calls our attention to the scale of the two proposals it reviewed, increasing the minimum wage to  $10.10 and to $9.00. The latter would raise the minimum wage to the level of bottom 10th percentile of “workers wages” (in $2013 dollars) and the former to the a level midway between the bottom 1oth percentile and the bottom 25th percentile wage levels. 

The chart also reveals that real wages at and near the bottom have remained stagnant since the 1970s, and have fallen significantly since the start of the Great Recession in 2009. And it shows that the minimum wage has remained below the 10th percentile median wage, barely staying at roughly the same inflation-adjusted level for about 25 years.  

This says nothing, however, about overall growth, to which aggregate employment is closely tied. On that score, CBO forecasting is inherently problematic. Note that CBO projects that real hourly wages at the bottom will grow from here on out. But we know that cannot happen with inequality continuing to rise, without (a) a significant increase in unemployment and/or a significant reduction in per capita hours worked, or (b) a major increase in aggregate growth. Aggregate growth is not increasing, which is ostensibly why Greta Brawner devoted here show to discussing growth: Indeed, she reported a significant decline of first quarter growth to an annual rate of 0.1%, which was shrugged off as the result of a cold winter.

These are broad implications of CBO forecasting that are unrelated to any changes in the minimum wage. CBO has never, so far as I know, considered the impacts of income and wealth redistribution on growth, and failure to consider those impacts seriously compromises CBO forecasting. We now know that income growth after WW II, a period of declining inequality, was far more robust than since since 1979. This graph, for example, compares growth rates for per capita national income, average bottom 90% household income, and top 1% income, between two consecutive 30-year periods, 1946-1976 and 1976-2006.  


It is important to note that overall national income growth (the yellow bar) was about one-third lower when income inequality was growing. Income inequality has continued to grow, and its growth has accelerated, since the Crash of 2008. These facts make the CBO projection of rising wage rates at the very bottom of the bottom 90% (the blue bar) wildly unrealistic. Realistically expected aggregate growth cannot sustain such averages at the bottom without, as I indicated, a major reduction in hours worked by the bottom 10% and bottom 25%, and that implies greatly increasing unemployment.

Since there is unrecognized unemployment growth built into CBO modeling, it is difficult to see how much credibility we can give to the CBO claim that increasing the minimum wage all by itself will cause a major reduction in employment. The best we can do is to try to understand CBO’s reasoning process.

The CBO Reasoning 

The report opened with a discussion of the isolated effects the CBO expects the minimum wage to have on the incomes of low-wage workers:

Increasing the minimum wage would have two principal effects on low-wage workers. Most of them would receive higher pay that would increase their family’s income, and some of those families would see their income rise above the federal poverty threshold. But some jobs for low-wage workers would probably be eliminated, the income of most workers who became jobless would fall substantially, and the share of low-wage workers who were employed would probably fall slightly (p. 1).

CBO then estimated that the proposed $10.10 minimum wage would increase the earnings of low-wage workers by $31 billion. (p. 2)

Beyond that, however, CBO merely listed a series of apparently subjectively evaluated factors, then jumped directly to its findings, without revealing its approach to forecasting or explaining how it reached its conclusions. CBO began its lengthy discussion (pp. 6-8) of how increases in the minimum wage affect employment and family income with this:

According to conventional economic analysis, increasing the minimum wage reduces employment in two ways. First, higher wages increase the cost to employers of producing goods and services. The employers pass some of those increased costs on to consumers in the form of higher prices, and those higher prices, in turn, lead the consumers to purchase fewer of the goods and services. The employers consequently produce fewer goods and services, so they hire fewer workers. That is known as a scale effect, and it reduces employment among both low-wage workers and higher-wage workers.

Second, a minimum-wage increase raises the cost of low wage workers relative to other inputs that employers use to produce goods and services, such as machines, technology, and more productive higher-wage workers. Some employers respond by reducing their use of low-wage workers and shifting toward those other inputs. That is
known as a substitution effect, and it reduces employment among low-wage workers but increases it among higher-wage workers. (p. 6)

Beyond the potential for scale and substitution effects to reduce output and employment, CBO mentioned that employers’ costs could be aggravated, following a minimum wage increase, by attempts to preserve wage differentials above the minimum wage, and that higher wage rates can be implicated by collective bargaining agreements that tie wage increases to the federal minimum wage.

CBO did not entirely dismiss the increased demand generated by $31 billion of additional income, but its conclusions were guarded:  

An increase in the minimum wage also affects the employment of low-wage workers in the short term through changes in the economy-wide demand for goods and services. A higher minimum wage shifts income from higher-wage consumers and business owners to low-wage workers. Because those low-wage workers tend to spend a larger fraction of their earnings, some firms see increased demand for their goods and services, boosting the employment of low-wage workers and higher-wage workers alike. That effect is larger when the economy is weaker, and it is larger in regions of the country where the economy is weaker. (p. 7)


All of the factors CBO mentioned are relevant, but there is no coherent evaluation of their importance, and CBO left the impression that it was rationalizing subjective conclusions that lack firm support. The report conceded that its conclusions were controversial: “There is a wide range of views among economists about the merits of the conventional analysis.” (p. 6) Here are some problems I see with CBO’s reasoning:

1. One would expect some economy-wide reduction purchases of goods and services whose prices have been raised in response to a minimum wage increase. However, there are reasons to doubt the significance of any “scale” effect on employment:

a. The demand for many goods and services is price-inelastic, and increasing prices in these instances do not cause much or any decline in output and employment;

b. Many firms enjoying profitable sales of goods and services the demand for which is more price-elastic, may elect not to increase their prices at all, in the interest of maintaining or expanding their successful share or control of markets. Such firms would absorb a reduction in profits, expecting profit margins to fall anyway following a decline in sales if they did raise their prices.

2. That substitution effects in the “production function” might cause significant unemployment is a more speculative proposition. There are many factors involved in the feasibility of substitutions, and it does not seem likely that in many instances an increase in the cost of a firm’s cheapest labor would suddenly make substitution profitable or feasible:

a. The marginal cost and productivity of all factors of production are relevant, and it seems unlikely that the functions performed by the lowest-paid employees can often be more cheaply performed by making them more capital-intensive and less labor-intensive;

b. Any such cost-saving innovations would likely have already been introduced, and would not need to await the provocation of a minimum wage increase;

3. There is insufficient discussion of the demand-increasing effect of the wage increases:

a. The $30 billion of increased demand produced by a higher minimum wage would be immediate and significant;

b. The assumption that a minimum wage increase is offset by reductions in higher-level wages is speculative and unjustified. As already discussed, a minimum wage increase is more likely in most instances to simply reduce profits, preserving its stimulating effect;

c. This is a static assessment, giving no consideration to multiplier effects and the potential for additional investment and lasting improvement of long-term growth;

d. Admittedly, prospects for long-term growth are more likely from direct spending through government programs and investment (financed by progressive taxation), but there is no reason subjectively to conclude that raising the minimum wage will necessarily cause employment to decline at all, much less materially.


These are my main concerns in connection with CBO’s analysis and conclusions, and I’m sure that experts on the minimum wage would likely provide a deeper analysis of its implications. It is fair to say, though, that the CBO’s unexplained analysis is too speculative to be legitimately termed a “forecast,” and CBO’s forecasts are seriously compromised in any case by its use of supply-side models and its failure to account for the depressing effects of rising income inequality.

Reducing inequality will promote growth, and increasing the minimum wage seems likely to have a relatively small but potentially long-term effect of reducing inequality, or slowing its growth. The CBO report fails to demonstrate that a higher minimum wage of $10.10 will cause the loss of up to one million jobs, and it is not clear that it will reduce employment at all. People who are looking for an increase in the minimum wage to substantially reduce inequality, however, are overly optimistic.

Although about $30 billion of additional income would be added at the bottom via an increase in the minimum wage, according to CBO, some $400-700 billion of wealth is moving to the top annually through the normal channels of income and wealth redistribution, a process that affects everyone beneath the top 1% income classification, and has already wiped out most of the former middle class. Clearly, benefits from an increase in the minimum wage are dwarfed by the impact of increasing income and wealth inequality. There can be no long-term correction of inequality or its stunted growth until the problem is rooted out at its source, by reinstating progressive taxation.

I have seen no indication that CBO understands the importance of demand to economic growth. The basis for such an understanding will be more fully addressed in my future posts dealing with the legacy of Gardner Ackley.

JMH — 5/4/2014 (ed. 5/5/2014)

This entry was posted in - FEATURED POSTS -, - MOST RECENT POSTS -, Economics, Wealth and Income Inequality. Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s