by on February 22, 2006


I suspect that this will fall on deaf ears, but both facts and logic are needed.

The statement: “Widespread visible homelessness emerged throughout America in 1982 because people could no longer obtain housing they could afford,” is neither an explanation, nor is it usefully predictive.

1. The homeless have been with us from the beginning of the United States, but there is not much point in examining the past beyond 1900. This is because data earlier than about 1900 are more anecdotal and less reliable, and because important trends had emerged by 1900 that still impact us today. Furthermore, I shall skip the period from the beginning of the Southwestern dust bowl [1921] until the end of World War II [1945] because the great migration, depression, and war are much more complex events that require additional consideration.

2. The most important information is contained in two trends in American pay statistics from 1900 to the present [These trends were reported in the New York Times during the late 1980s, but the best present-day source is obtained by downloading the complete .pdf file at http://www.frbatlanta.org/invoke.cfm?objectid=761A075D-E26C-A7B7-D54E7D4438315F59&method=display

Income and Education of the States of the United States: 1840-2000. Federal Reserve Bank Working Paper Series, Federal Reserve Bank of Atlanta. Searching “income by education by year” in Google Scholar or in FirstGOV will yield more papers and analyses.

The first trend is a reported steady increase in the pay [corrected for inflation] of working Americans in all sectors of the economy that, excluding the period 1921-1945], ran from 1900 to 1980. In fact, if you look at the labor statistics graphically and draw the trend line from 1900 to 1980, the pick-up in post-war salaries is just about where it should be if the trend had continued on a straight line through the bumpy period of the great depression and war. Although the slope of each salary trend line is different for ethnic, social, educational, and employment sector subsets of the population, even the lowest slope shows a long rise from 1900 to 1980.

3. The second trend in these labor data became clearly evident around 1981 for the subset of workers who did not complete high school [see ref. page 14]. For those workers, in 1981 the long 20th-century rise in incomes ceased. The averaged salary line for them, viewed from today, actually flattened out and began to drop [relative to inflation]. A couple of years later, the subset of workers who completed high school experienced the same flattening of incomes, with eventual downtrend. In fact, this was a reversal of the long salary rise for almost every education group. The last group to show this change, in the 1990s, was the group of workers who had completed university and attempted graduate studies. So far, the group of workers who have completed post-graduate [MS, PhD, MD, etc.] degrees has shown some tapering in its increase, but has not yet flattened or turned down.

In a specific example: In 1900, most presidents of American companies had not completed college. None had advanced degrees (advanced education was, in fact, thought damaging to the business mind). By the mid-1990s more that 50% of presidents or CEOs of American companies had doctoral degrees. In technology, pharmaceutical, and biotechnology sectors, the percent of PhDs and MDs is much higher. Such trends are found throughout American business and academia.

Conclusion from these data: In about 1981 the previous long 20th-century rise in worker incomes began to flatten out and drop, progressively by level of education. Higher education delayed this halt, but did not stop it. Closer examination shows that only high-levels of technical education have, on average, exempted workers from this trend. Specifically, at any job level, worker qualification, which began an upward rise in the 1970s, began to increase rapidly in the 1980s and has continued to increase since.

4. Coincident with this trend [and probably causative] is the trend toward job loss through automation. Although G. E. Moore proposed what is now called Moore’s law in 1964, the progressive doubling of the amount or memory or processor speed in each 18-month period really began to kick in during the late 1970s and early 1980s. See: http://en.wikipedia.org/wiki/Moore’s_law

This was because processor size and memory density had, by around 1980, reached a level practical for wide-scale industrial automation. When automation was slow and cumbersome during the 1960s, because large and very expensive machines were required and only very simple tasks could be automated, by the 1980s automation could sit on a desk top and displace desk workers. Indeed, this happened. As late as 1980, secretarial training was a viable career option for many young women and some young men. A person with good secretarial skills, including shorthand, was assured of a lifetime of adequate to good pay.

By 1990, the job of secretary no longer existed and typing pools were a thing of the past. When an executive has a secretary today, it is for prestige, not for necessity.

This has happened throughout American Industry. For example, in 1980 employment in the city of Milwaukee, at manufacturers such as Briggs & Stratton, who make small engines, commanded production-line salaries of $40,000 to $45,000 per year. During the 1980s Briggs & Stratton moved some plants to Mexico, but, more importantly, automated the assembly line. Now the folks making the equivalent year 2005 salaries are the production-line programmers [very many fewer than production-line workers]. Milwaukee went through a local recession in the late 1980s in which families used to joint salaries of around $80,000 per year suddenly found themselves subsisting on Tupperware sales and flippin’ burgers at McDonalds…!

Conclusion from these data: Automation [and outsourcing] reduced jobs, particularly in the previously high-paying manufacturing sector. These workers were displaced downward, in a cycle that increased the required education, skill level, and responsibility-requirement at every job level. As this process continued, workers with the least education, the lowest skill levels, and greatest inclination toward irresponsibility fell off the job tree. This substantially increased the percentage of permanent unemployed [or marginally employeed] workers.

Two consequences flowed from this:

First, people in the lower and less enduring rungs of the job ladder became more and more prone to abrupt periods of unemployment and consequent homelessness. This was not only because low-level jobs often simply disappeared, but also because such workers made too little to save for the resulting layoffs and because poorly-paid workers rarely have medical insurance and often find themselves unemployed because of illness.

Second, persons with emotional problems, problems of substance or alcohol abuse, or simply attitudinal problems were increasingly at risk of continued and long-term unemployment. The casual labor or part-time work that had previously sustained such people was now filled by much more stable workers who had been displaced downward from better jobs.

5. Finally, beginning in the late 1960s, two trends vastly changed our treatment of the mentally disabled throughout the United States. Again, from 1900 until around 1970, the institutional treatment of mentally disabled individuals improved steadily [the rate and quality are much more subjective than salaries]. The first trend was in mandatory institutional care. during the 1960s a number of books and films showed the plight of individuals trapped in public and private mental institutions. Nurse Ratchett became symbolic of heartless institutionalization of the mentally disabled. As a result of widespread liberal concern, self-release became increasingly easy and many persons who required medication to cope with life found themselves on the streets or in half-way houses. The second trend consisted of closing mental institutions by conservative politicians. The net result was that many folks who required constant medication to cope found themselves first in half-way houses, then on the streets. The private half-way house business boomed and entrepreneurial social workers found themselves teamed with property developers in an unholy feeding frenzy at the HUD trough.

Conclusion from these data: Substantially larger numbers of emotionally-disabled people flowed from institutional care through half-way houses and out onto the street. Such people frequently entered the cycle of mental illness, drug dependency, and alcoholism. This has swelled the numbers of homeless who are permanently caught in this cycle far beyond the levels of previous years. Estimates vary, but such people appear to constitute at least one-third to one-half of the homeless at any one time, but around 75% of the chronically homeless. This also constitute the most difficult group of homeless to help. People entering homelessness through simple unemployment have a high frequency of getting off the streets and into some more-or-less stable home within a matter of weeks. People remaining homeless for more than a matter of two or three weeks have an

increasing chance of being caught up in the cycle that perpetuates homelessness. An increasing number of the latter group are young people from suburban homes. Many of these are minors.

Yes, the rising real estate prices have played a role. This role has made the real trend worse, not caused it.

In San Francisco, both conservatives and “progressives” have resolutely refused to make this simple analysis part of their planning for homelessness. The conservatives willfully see all homeless as the chronic older homeless; the “progressives” willfully see all homeless as unfortunate single mothers. The complexities of this situation are lost in the idiotic debate over shelters vs. housing. “Care Not Cash” has been a good program for the chronically homeless caught up in the illness-drug-alcohol cycle; it does not do much for a single mother or a destitute family with two children. Immediate shelter for the single mother and child is great, unless it is also filled with drug users and alcoholics. The broke family has a high probability of getting out of the shelter and into self-selected housing. The chronically homeless need special care and benefit from long-term housing. No one benefits much from the Welfare Developers.

I had hoped that we would see a clear plan being implemented to perform “triage” on the homeless, providing appropriate solutions, help, and either shelter or housing. The “Shelters vs. Housing” screed suggests we have not gotten there yet. The folks who draw their incomes from the problem remaining unsolved still rule.

Wayne Lanier
San Francisco

You can submit letters to the editor by clicking on this link: rshaw@beyondchron.org
or by writing to:
Beyond Chron
126 Hyde Street
San Francisco, CA 94102
415-771-9850 (phone)
415-771-1287 (fax)

Filed under: Archive