Years of Poverty, Years of Plenty, The Changing Economic Fortunes of American Workers and Families

Years of Poverty, Years of Plenty,
The Changing Economic Fortunes of American Workers and Families
by Greg J. Duncan with Richard D. Coe, Mary E. Corcoran, Martha S. Hill, Saul D. Hoffman and James N. Morgan

Institute for Social Research, The University of Michigan, Ann Arbor MI 1984
Cost and standard of living - United States - Longitudinal Studies.
Income distribution - United States - Longitudinal studies. Labor and laboring lasses - United States - Longitudinal studies.
ISBN 0-87944-285-9 clothbound
ISBN 0-87944-289-0 paperbound

If I told you that "the poverty rate in America is 13%", you should think that the problems of poverty are quite enormous. On the other hand if I said "only .7% of the population are persistently poor every year", your reaction should be quite different. The fact that both statistics are correct, shows the need to distinguish between "longitudinal studies" and "cross sectional studies." A census or cross sectional study tells you that "x" percent of the people are poor but tells me nothing about whether they are poor for a week, a month, a year or 10 years. A longitudinal study follows the same people year after year. We can have our debate over the appropriate remedy only after we know more accurately what we are talking about. If you don’t know the difference between bouts of poverty and persistent poverty you are not in the conversation.

Greg J. Duncan’s book Years of Poverty - Years of Plenty is an analysis of a longitudinal study called the Panel Study of Income Dynamics that started in 1968 and examined the lives of 5000 American families/individuals over the course of a decade. The outcomes of longitudinal studies changes your perspective about conventional wisdom because "helping the poor" becomes a moving target. Another statistical lesson to be learned is watch out for people spouting "family" data. It is no surprise that the makeup of families change substantially over time. Marriage, children, death, divorce, promotions, layoffs, emancipation, retirement, etc all mean that a statistical analysis of "a family" is virtually impossible. Keep your eye on the individual. By using the Panel Study data, Duncan reviews economic mobility, poverty, welfare, work, black men and women’s income. The importance of this book, is to examine, in a non-technical fashion, people’s lives over time to reveal the nature of poverty.

Although the book is packed with many different facts and topics, I found five main lessons concerning poverty: 1) economic mobility is alive and well; 2) many people experience short spells of poverty; 3) the persistently poor are few in number and different demographically from the general population; 4) the absence of persistently elderly poor raise interesting policy questions, and 5) persistent welfare use was very rare from 1968 to 1978.

Economic mobility is alive and well. Those who are poor today are, as a general proposition, not likely to be poor over time. People grow up, gain work experience, receive on the job training and over time, their incomes go up. One third to one half of all who are poor this year are NOT POOR THE NEXT. This is an amazing statistic all by itself. This is not India. We do not have an untouchable class. If someone says we must "help the poor," who are they going to help? Here today, gone tomorrow. The study noted that "... less than one in a hundred (0.7 percent) was poor all ten years." (P41)

The second lesson is that the experience of poverty is actually widespread. 25% of the population have at least one experience with poverty during the 10 year period. Bad stuff happens to lots of people. For the most part, they handle it and move on.

The third lesson is that persistent poverty, that is, people who are poor year after year is quite small. You can pick your own favorite statistic. Only .7% of the population were poor absolutely every year for the 10 year study. 2.6% of the population were poor in 8 of the 10 years. (Duncan settled on this statistic as his definition of persistent poverty.) 5.4% of the population were poor in 5 or more of the 10 years. The main point is that these numbers are quite different from the 13% poverty rate publicized from the cross sectional census statistics. The characteristics of those who are persistently poor are also very different from the general population. Those who experience an occasional bout of poverty are virtually identical to the general population. "Persistent poverty is heavily concentrated among blacks, and particularly among families headed by a black woman." (P50) Blacks are 12% of the population and 62% of the persistently poor. Only 19 percent of families are headed by black women but they constitute 61% of the persistently poor families. Only 30% of the population live in the South, yet 68% of those who are persistently poor live there. 4 out of 5 persistently poor people live in a towns smaller than 500,000 people.

Fourth, I wish the book had distinguished between how many elderly would have been persistently poor with and without social security. It does mention that social security "does not cover all of the elderly" and that there is low participation in Supplemental Security Income programs (only half of those eligible). What we saw above is that approximately 2.6% of the whole population are persistently poor but only one-third of those are elderly, i.e. .866%. Milton Freedman in Capitalism and Freedom asked whether mandatory social security was justified. He correctly noted that some people, if given the option, would choose to live for the moment and spend all of their money. Those people might later become burdens to the rest of us if their families did not take care of them. Freedman suggested that the argument be resolved by statistical analysis. If 90% of the elderly actually became a burden to society, then the mandatory aspects of social security would be justified. If, on the on the other hand, only 1% of the population became a burden, how can we justify depriving 99% of the population the freedom to manage their own retirement funds. I am not suggesting that Freedman’s statistical question has been answered by Years of Poverty, but the fact that less than one percent of the population is both elderly and persistently poor is certainly cause for further study.
Lastly, welfare is always a hot topic of discussion. Welfare programs provided assistance to 25% of the population at least once during the 10 year study. Very few of those people stayed on welfare very long and very few were completely dependent upon welfare for their support. About 2% of the entire population were persistently dependent upon welfare. They were "disproportionately female, black, and have children in the home." (P92) From these welfare statistics, Duncan concludes that the program is not guilty of creating wide spread dependency. While it may be true that welfare is not a deadly disease that once touched becomes permanent, it does appear to be having a serious impact upon a small group of citizens. It is amazing how different people respond to the same statistical analysis. Duncan sees the widespread mobility and concludes that welfare isn’t so bad, because it didn’t cause large numbers of people to remain persistently poor and dependent. I look at the same mobility statistics and ask whether we need welfare if most people are able to move up and out of poverty so quickly. What is wrong with private or charitable safety nets for those who experience a short bout of poverty? Why can’t people save money or fall back on family for brief periods of time? Obviously when you are poor, you can’t save, but remember that for most people, poverty bouts are very brief. For most of the time, people are fairly capable of normal monthly saving. As Edgar K. Browning points out in Stealing From Each Other, 46 per cent of the poor are in the process of purchasing their own home. This issue reveals one of the inadequacies of this Panel Study. They did not inquire about private safety nets such as saving, investment, home ownership and association with charitable institutions.

During the Great Depression this country was sold on the proposition that our problems were too big for a system of voluntary charitable assistance to the poor. As a result we now have a large welfare system and a nearly universal Social Security Administration. It would appear that we should have these debates over again based upon facts from longitudinal studies. Persistent welfare use at 2% of the population may have been actually been reduced by the Clinton era reforms. The tail of the Great Depression is wagging a very large dog.