Poverty, Education, and the Fallacy of Composition

It is true of each person at a concert that he or she can walk into the auditoreum's single door at precisely 8 p.m., but it is not therefore true that the entire audience can walk through the single door of the auditoreum at precisely 8 p.m. Each student in a class, I like to think, can with luck and hard work write the single best final exam, but it does not follow from this that all of the students can write the single best exam. Logicians have a name for the mistake in reasoning that consists in inferring, from the fact that a proposition is true of each member of a group, that the proposition is true of all the members of the group. They call it The Fallacy of Composition.

The belief, endlessly echoed and virtally universally believed, that education is the key to the elimination of poverty, rests on a very simple but seductive commission of the fallacy of composition.

Think of a company -- National Porta-Toilet Corporation, let us say -- in which there are a thousand jobs, ranging from President and CEO to mail room clerk. The jobs, we shall suppose, can be arranged pyramidally according to the salaries or wages associated with each position, with a small handful of top jobs carrying high salaries and lavish perks, a slightly larger number of upper management positions, more middle management slots, down to a goodly number of production jobs, secretarial jobs, and the like, with correspondingly lower salaries, and maybe not even health benefits or paid vacations.

If this is a typical American corporation, then it will almost certainly be the case that there is some level of educational credentials associated with each position. A high school diploma or equivalency may be required on the loading dock, a college degree for the middle management positions, and an MBA or other advanced degree for the top executive slots.

If an individual wishes to improve his or her chances of getting one of the favored jobs, therefore, a good strategy is to stay in, or go back to, school, and earn some more degrees. Leaving aside nepotism, bias against women and minorities, and other market distorting facts of American life, there is no question that getting more education [or, more precisely, more degrees -- not at all the same thing, of course] is a first-rate strategy for moving up the job pyramid. Those at the bottom of American society are, taking all in all, those with the scantiest educational credentials.

This strategy is also a good one for some relatively small sub-group of the National Porta-Toilet workforce -- African-Americans, say -- who, we may suppose, are disproportionately represented in the lower wage and salary levels. But it takes only a moment's thought to realized that this cannot possibly be a successful strategy for the ENTIRE workforce.

Imagine, if you can, that in a burst of focused ambition, every single worker below the level of upper management at National Porta-Toilet goes to night school and earns an MBA. Will the company now promote them all to senior management? Of course not. Who would then make the porta-toilets, load them onto trucks for delivery, file the invoices, and answer the phones?

Now think of the entire American economy as though it consisted of one vast corporation -- a sort of cross between GM, Boeing, Google, Microsoft, Archer Daniel Midlands and all, run wild -- a corporation with somewhat in excess of a hundred million employees, having both public and private sectors, and generating the totality of the goods and services produced in America.
Clearly, any individual wishing to make his or her way in this great mega-corporation will do well to get as many educational credentials as possible, but although that is a splendid strategy for an individual who wishes to make more money and get a better job, it cannot possibly be a national anti-poverty strategy for raising the entire bottom of the workforce so that all Americans are in the middle class. The reason is simple: the jobs exist BEFORE they are filled, and are defined, with wages and salaries and benefits associated, according to the operational needs of the processes of production and distribution of goods and services, not the other way around. If there are an unexpectedly large number of people with MBAs, this does not provoke an expansion of the ranks of upper management. So education cannot be the solution to poverty, save for some, and then only so long as others fail to pursue education.

Immediately, objections will arise.

The first objection is that it is perfectly possible for an entire nation to raise itself out of poverty through education, as witness the economic successes of some of the Asian tigers. But such success is possible only so long as there remain hundreds of millions of men and women elsewere in the world who can be consigned to the low-paying jobs that the successful nation has eschewed. The real meaning of globalization is that entire continents become the low-wage working class of the world system.

Because the world economic system is so large and complex, it is easy to imagine that the poverty wages of Africa or Latin America or Asia are a consequence of their inadequate educational attainments, and that every country in the world could undergo an economic miracle, with appropriate capital investment, a strong civil society, and universal education up through the tertiary level. But who then will do the low-paying jobs on which the more affluent depend for goods and services?

Neo-classical economic theory suggests a second objection. The assumption underlying the mathematics of that theory [the details, although quite lovely mathematically, need not trouble us here] is that each firm has an endless array of techniques available for the production of whatever goods or services it sells, from among which it chooses techniques according to their profitability at current market prices for inputs and labor. If the educational attainments of the labor force change [and assuming that with that change actually comes a change in the kinds of jobs workers are capable of performing], employers can shift to techniques of production that require a better educated labor force, and this, it is thought, will result in a flattening of the income pyramid. Instead of a GM with many manual laborers and relatively few "suits," one will move toward an economy of Microsofts and Googles. The bottom will be pulled up, and low wages will become more and more a thing of the past.

There is obviously a good deal of truth in this objection. The patterns of compensation of the work forces of information age companies are quite different from those of the old rust belt. And clearly, this upgrading of educational attainments has been going on for a very long time in America. The father of my first wife never finished high school, and yet he ended his career as a Vice President of Sears, Roebuck. Today, he would not make it into a management trainee program with less than a college degree. A hundred years ago, the American labor force was primarily agricultural. Today, fewer than 2 percent of the labor force provide food and fiber for the entire nation, and service jobs vastly outnumber production jobs. Techniques of production have been transformed [except in higher education, where I and my colleagues still use today the same basic pedagogical techniqes that my professor's professors used at the turn of the nineteenth century!]

We might therefore expect to see, over that time, a steady flattening of the pyramid of wages and salaries. The truth is completely the opposite. The total wealth produced by the American economy has soared, but the shape of the distribution has remained essentially unchanged. For a while, in the 70's and 80's, the pyramid flattened; lately, it has sharply steepened. But these changes have been almost entirely due to changes in federal tax policies and welfare programs, not to technique substitutions triggered by changes in the educational attainments of the labor force.

Think for a moment of the consequences of the dramatic layoffs and downsizings in the corporate world that turned the lives of so many employees upside down in the eighties and nineties. Companies laid off hundreds of thousands of middle managers and other workers with superb educational credentials. As a result, there was a surplus of well trained unemployed workers in the labor force. But this did NOT trigger a massive shift in choice of production technique. Instead, these workers by and large were forced to take less well compensated jobs in which, quite often, their skills and education were under-used. In that situation, getting more young people to stay in school and earn college degrees simply made the situation worse. To be sure, getting more educational credentials was still a good strategy for an individual -- indeed, in the face of the competition for the favored jobs, degrees, especially from well-known colleges, became ever more valuable assets. But education was not, and could not be, a systemic solution to the problem of poverty.

A little thought experiment, prompted by the insights of Karl Marx, might be helpful here. Instead of thinking about how much money people make and spend, think instead about how much of other people's labor time they consume. When I travel, I stay at hotels. Each day, my room is straightened up, my bed is made, and my bathroom is cleaned. [a quick read of Barbara Ehrenreich's great book, Nickled and Dimed, is useful here.] It is intuitively obvious that the women who perform the maid service in the hotels cannot possibly afford to stay there as customers.

An even more immediately obvious example is a full time household worker. Since it takes her a day's labor to earn what she needs [we may hope] for food, clothing, and shelter, she cannot possibly herself afford to hire a full time household worker to clean her own dwelling, look after her children, as she looks after her employer's children, and so forth.

Some elements of a household's "market basket,' as the Bureau of Labor Statistics calls the array of goods and services we purchase, can be made to require less and labor for their production by machines. This is why even poor people in the United States can afford decently made clothing and television sets. But health care, for example, which is labor intensive, is costly in labor time, and hence difficult for poor people to afford.

In short, even in an America of superbly educated men and women, as things now are, the comfort and convenience of the well-to-do can only be secured by the poverty of those who provide important elements of that comfort and convenience.

What can be done? Is the Good Book right? Are the poor always with us, no matter what we do?

That is a subject for another post.