Sociology 498: Homelessness

Jencks: The Homeless

Overview

Jencks's book began as a book review (of Burt and Rossi) for the New York Review of Books. By the time he had finished writing that review, he had enough information (and interest) to write his own book. This is perhaps my favorite book on homelessness (see my top ten reasons for liking this book!). His book is short but full. He reviews all the major explanations that have been given for homelessness and sorts out the evidence for and against each. Jencks is especially good at cautioning us not to conclude too much from any one study or one piece of evidence, and yet not to avoid drawing some conclusions, however tentative.

Jencks's research strategy is also different from the others. Like Burt, he concentrates on macro-level questions but instead of comparing cities, Jencks compares years. Why was there more homelessness in the 1980s than in the 1970s? What changed about the country that gave rise to this new phenomenon?

Jencks's preface and his table of contents will help you understand where he is going with his causal analysis. Read them carefully. Comparing them to our list of major causes you will find that some of our causes (e.g., housing) are listed in both his good explanations and his bad explanations. This makes his analysis more complicated than what we have read so far.

Chapter 1: Counting the Homeless

While some people are easily defined as homeless, there are many border-line groups who may or may not be defined as among the homeless. Identify these border-line types and Jencks' decision to include or exclude them from his definition of the homeless. How does this differ from our previous analyses? Do you agree with Jencks' definition?

Chapter 2: Estimating the Increase

Despite the difficulties in counting the homeless, Jencks perseveres in creating estimates at different points during the 1980s. What is his conclusion about the trend? What is the difference in levels between the 1950s and the 1980s? This trend is the main fact (his "dependent variable") that he will spend the rest of the book trying to explain.

Chapter 3: Emptying the Back Wards

We have been critical of the common belief that mental illness causes homelessness. Jencks nicely reviews most of our objections and then tells us not to make such a big deal out of these objections. He ends up defending Rossi against our criticisms and concluding that de-institutionalization has been an important cause of higher rates of homelessness. Where do you stand on this issue after reading Jencks?

What is the "awkward fact" that Jencks has to get around in arguing that de-institutionalizing the mentally ill is one cause of the trend in homelessness? How does he try to get around that problem? Why does he spend so much time reviewing the history (the five waves) of de-institutionalization?

Chapter 4: The Crack Epidemic

Jencks closes the chapter with a comparison of liberal and conservative explanations of homelessness. He argues both are correct. Another way of thinking about this is our macro-micro distinction. How does the macro-micro distinction (and the liberal-conservative distinction) help us assign responsibility (blame?) for the rise in homelessness.

Chapter 5: Jobs and Marriage

In this chapter, Jencks combines two items from our list of causes. In part, he divides the two explanations by gender: lack of jobs may explain men's homelessness; marital breakdown may explain women's homelessness. (See the second paragraph). But Jencks also tries to combine the two explanations, much as Rossi did in his section on "The Accumulation of Troubles". What does Jencks mean by his "vulnerability index" (p.50)? The argument in this (and later) chapters gets complicated because he is trying to combine arguments that we have been trying to keep separate and evaluate separately. Separating the arguments may be good for simplicity, but it may hide the real process of becoming homeless. The argument gets complicated in this chapter, but that is because Jencks believes that the causes of homelessness have to be treated jointly.

Most of the arguments in this chapter derive from the numbers in Tables 5 and 6. Each row tells a slightly different story. The chapter works its way through these tables almost row-by-row. Be prepared to follow Jencks through multiple zigs and zags of an argument. Often he starts with the basic overall plausible argument, often with some evidence that supports the argument. Then, he will usually point out what may be wrong with that evidence, sometimes citing previous writers on homelessness. Then, he will rebut those criticisms. This may even go on for another round or two before he comes to his conclusion. If you can sort out these various zigs and zags of the argument, you will appreciate what Jencks is doing much better.

Jencks begins with evidence about "long-term joblessness". First, how is this different from "unemployment"? How has long-term joblessness changed over time? and does this support an economic explanation for homelessness or not?

Jencks then gets diverted (pages 52-53) into an excursion about the possible causes of the changes in long-term joblessness. This is interesting, but not logically connected to the main task of exploring the link between joblessness and homelessness. Nevertheless, realizing that this is a diversion, what is Jencks's evidence concerning the cultural explanation of the rise of joblessness? What is Jencks's evidence concerning the economic explanation of the rise of joblessness?

Joblessness creates homelessness only if the jobless fall through the "safety net". What are the three elements of Jencks's safety net? Then, Jencks tries to evaluate whether the holes in the safety net grew in the 1980s. How does the evidence about personal income help evaluate changes in the safety net? What is Jencks's conclusion?

How does evidence about marriage help evaluate changes in the safety net? What evidence is consistent with the marriage breakdown idea? What evidence is inconsistent with the marriage breakdown idea? (i.e., make sure you follow the zigs and zags of this argument!)

For women, Jencks begins by evaluating economic changes. Are these consistent with an economic explanation of the rise of women's homelessness?

Then, he evaluates marriage and family changes. What part of these changes is consistent with, and what part is inconsistent with the rise in homelessness?

Then Jencks gets diverted again into an interesting (but not essential) speculation about several possible causes of the decline in marriage (from the bottom of page 56 to the end of this section on page 58). What are the causal arguments? and what is the evidence about the arguments?

Finally, table 6 tries to pull all these factors together to estimate changes in the numbers of people most vulnerable to homelessness. Some of the changes seem to argue for a joint economic/marriage causal explanation. Which ones? But there is another "inconvenient fact" in this table that very much weakens this causal argument. What is the problem with the numbers in table 6 for an economic/marriage explanation?

Chapter 7: Social Skills and Family Ties

Jencks is not at all clear about this, but you can dig out from this chapter the time trend in adults living alone. Is this consistent with a rise in homelessness in the 1980s? That is, does living alone increase as homelessness increases?

Families often make the difference for poor adults in whether they become homeless or not. Why, according to Jencks, is this not a very good explanation of the rise of homelessness in the 1980s? What is the evidence?

Chapter 6: The Destruction of Skid Row

I think it may be easier for you if you know right away where he is headed. His principal conclusion is summarized in the next to last paragraph of the chapter on page 74 and in the third paragraph ("My best guess...") on page 67. Here he argues that it is the combination of housing shortages and increases in poverty that led to the rise of homelessness in the 1980s.

The first section asks what happened to the number of single-room rental units. Simplify this: when did single room housing decline? in the 1970s? in the 1980s? Why does Jencks conclude that the timing of the decline causes a problem for housing shortage explanations?

If the number of single-room units doesn't explain the rise of homelessness, maybe a rise in the price of minimal housing would help. What is the time trend in very low rent rooms? Is this consistent with a housing shortage explanation? What is the change in rental prices for 1 room units? Is this increase in price a result of housing shortages as Rossi originally claimed?

Third, Jencks tries to ask why the homeless may make a choice not to spend their money on a room in 1980 but might have made a different choice in 1950. What affects that choice? Is thinking about choices a good way to think about homelessness? Do the homeless really have choices to make or are they in trouble precisely because they don't have the resources needed to make choices?

Finally, Jencks asks why new cheap housing didn't spring up to meet the rising demand for cheap housing that resulted from the rise in poverty? This pushes the causal chain back one step: what caused the shortage of housing? (rather than did a shortage of housing cause the rise of homelessness?) Who are the "villains" of this market failure? As usual, Jencks finds there is alot of blame to spread around.

Chapter 8: Changes in the Housing Market

Some conservatives have singled out William Tucker's attack on rent control as one of the causes of rising homelessness. This can be a tricky argument but it's worth understanding. Jencks attacks this reasoning not because the argument is necessarily wrong, but because it doesn't fit the evidence Tucker shows that cities with more rent control have more homelessness. There are at least three problems with the evidence that don't support a causal conclusion here. Make sure you understand these problems.

Chapter 10: Do Shelters Cause Homelessness?

If people are choosing to become homeless, does it follow that if we make the homeless situation less intolerable by creating more humane shelters, will we actually increase homelessness? That is, are we likely to make the problem bigger by making it nicer?

Chapter 11: Some Partial Solutions

Jencks calls these partial solutions because different causes (see the opening paragraph of the previous chapter) require different solutions. He divides the causes into three different populations:

• mothers with children
• single adults
• mentally ill