Warren and Tyagi argue that, in general, middle-income families do not spend wastefully beyond their means. They support their thesis with data on consumer expenditures that show, they maintain, that spending on necessities has risen as rapidly as income, leaving little income available for “frivolous spending.”

I agree that most middle-income households consume prudently and that their spending on necessities has been increasing; however, I am not convinced that this spending has been rising as rapidly as personal income. The authors focus on four-person families, but these represent a shrinking minority of the U.S. population: by 2002, four-or-more-person households made up less than one quarter of all households. Some evidence shows that households without children are in fact the ones most likely to “over-consume.”

The authors suggest that middle-income families have had little choice about allocating an increasing proportion of their incomes to transportation and housing. But it is not clear that consumers need the more than 30 million recent-model passenger vehicles that they purchase each year instead of older models. And Americans have recently spent more on housing in part to take advantage of expected future home appreciation, despite the availability of relatively less-expensive rental housing. Significant increases in entertainment spending by all middle-income groups provide additional support for the view that their spending on wants, not needs, has been rising more rapidly than personal income.

If the authors had examined changes in wealth as well as income, they would have observed that, for middle-income households, available net financial resources have been growing. For several decades, the typical (median) household’s net wealth (as measured by the Federal Reserve Board) has been increasing; between 1992 and 2001 alone, it increased by 40 percent in constant dollars. While this expansion has been inflated by the replacement of defined-benefits plans, which are not counted in wealth figures, by defined-contribution plans, which are, the point here is that the financial resources available to middle-income households have been growing, suggesting that most middle-income households still have the capacity to afford any increases in necessary spending in the present—though not, unfortunately, during what many hope will be their retirement years.

In their research, the authors were limited by inadequate theory and data. In any analysis, it is not easy to distinguish between necessary and discretionary spending. In part, this is because the two concepts are social as well as economic. For example, even ten years ago, few considered cell phones and computers to be necessary. Furthermore, exactly what portion of many essential goods and services—say, automobiles—would one consider to be necessary and what portion frivolous? Even if we could reach agreement about these distinctions, we would have difficulty measuring them. Just ask the many economists who try to measure product improvements in order to ensure accurate price indexes.

In assessing affluence, I think it worthwhile to consider the attitudes of individuals about their own financial condition as well as data about the conditions themselves. Having just reviewed several years of relevant survey data, I have found that about two thirds of U.S. households consider their financial condition to be excellent or good, while the remaining one third think their condition is fair or poor. Not surprisingly, these responses are highly correlated with income, which suggests that the percentages apply to middle-income families as well as to all households. One would think that those who believe their financial condition to be excellent or good would have discretionary financial resources available.

In sum, most middle-income households are perhaps not as financially constrained as the authors suggest.

The authors begin and end their essay by discussing bankruptcy reform. In fact, their outrage at this reform (which I share) may have inspired their analysis debunking the “over-consumption myth.” However, the most effective argument of “reformers” was that most Americans were financially prudent and thus would remain unaffected by the new bankruptcy restrictions. These restrictions were, instead, intended to restrain the irresponsible spending and debt accumulation of a relatively small minority. By making this distinction between the responsible majority and the irresponsible minority, creditors successfully turned bankruptcy reform into the consumer equivalent of welfare reform.

Was this minority in fact irresponsible? In earlier research, the authors found, for example, that a high percentage of personal bankruptcies were associated with unpaid medical debts. It is important to add that these debts were so damaging because bankrupt people were also frequently saddled with large credit-card debts and held few, if any, financial assets. Moreover, since most bankrupt households had low- or lower-middle incomes, they were less able to draw on the resources of extended family members, who tend to be part of the same income classes, than were upper- and upper-middle-income households that became financially insolvent.

Couldn’t these people who became bankrupt have built wealth instead of debt before they experienced the income losses or unexpected expenses that drove many into bankruptcy? After all, some less-affluent Americans have done so. On the other hand, because of fundamental changes in the American marketplace, it has become increasingly difficult for consumers to build savings and avoid unsustainable debts.

For most low- and lower-middle-income Americans, there are few attractive savings options, yet much opportunity and encouragement to accumulate debt. For regular savings accounts, the starting point for most less-affluent savers, most banks require ever-higher minimums and pay yields lower than inflation rates. An alternative to these accounts used to be U.S. Savings Bonds, which were available at banks and credit unions, but purchases of these bonds can now only be made online. Most importantly, only a minority of lower-income employees have access to a retirement program at work.

Meanwhile, unpaid credit-card balances represent the most rapidly growing form of consumer debt. This high-priced debt is so attractive to creditors that they mail out more than five billion credit-card solicitations annually, extend more than $3 trillion in card-related lines of credit, and, until recently, allowed cardholders to make minimum payments of only two percent. With such opportunity and encouragement, combined with expanding marketing of goods and services that can be financed, is it any wonder that Americans now owe nearly $800 billion in credit-card debt?

So to what extent are creditors and consumers each responsible for unsustainable consumer debt levels? An analogy may be relevant here: suppose distillers mailed dozens of liquor samples annually to almost all American households, and as a result, ten to 15 percent of these families experienced serious drinking problems. Would anyone not consider the liquor companies wholly or largely responsible for this excessive drinking and its related problems? Similarly, we should be critical of those lenders who not only unsuitably extend credit but also have had the chutzpah to successfully lobby Congress for bankruptcy restrictions that will likely inflate their already-high profit levels. Tragically, as the authors demonstrate in other research, these restrictions will harm many Americans who acted prudently yet were driven into insolvency by factors beyond their control.