Can Government Build Cities?
A disturbing trend after Katrina was summed up in George Bush's promise to have the federal government completely rebuild the Gulf Coast better than before the storm, and do so with taxpayer money. Can we really expect government to create quality cities using redistribution, government programs, and regulations?
The Bay Area Center for Voting Research has published a list of America's Most Liberal Cities. By "liberal," the Bay Area Center for Voting Research means the contemporary, American political definition of the word, which involves a willingness to use the taxing, spending, and regulatory powers of government to redistribute wealth and to control behavior.
This is a much different thing than classical liberalism. Classical liberals, such as Ludwig von Mises, reject contemporary, American "liberalism." While I myself normally avoid words like "liberal" and "conservative," because they do not clearly identify whether someone is pro-government or pro-individual liberty, I will use these words in this essay because this is the terminology of the Bay Area Center for Voting Research.
At the top of the list of America's Most Liberal Cities is Detroit, Michigan. Detroit, with half the population it once had, and with a quarter of its land vacant or abandoned, is indeed a monument to liberalism. Although the city is financially bankrupt, it is able to find big bucks to subsidize marquee events and to underwrite billion dollar giveaways to professional sports teams. Its unemployment rate may be comparable to that of an east German lander, but its jet-setting, wheeler-dealer mayor has an unlimited expense account. And, whoever is the statewide Democratic candidate, he or she can count on Detroit, where election officials may be slow in tabulating the vote, but they are sure to come up with enough of a margin to make the difference.
Of course, just because America's Most Liberal City is a disastrous combination of stifling taxes, high crime, poor schools and bad roads doesn't necessarily mean that Liberal Cities tend to be that way. Detroit could be a fluke at the city level, the same way that North Korea and Zimbabwe supposedly are flukes at the national level. Just because people are reduced to eating grass in some places where totalitarian socialism rules, doesn't prove that totalitarian socialism tends to impoverish a nation. And, just because some cities that tend to vote liberal become dysfunctional, doesn't prove that liberalism tends to make cities dysfunctional. Examples are merely illustrative. Empirical proof requires the analysis of a statistically-valid sample, and theoretical proof requires a strong connection of cause and effect.
Liberalism and Unemployment
Being rather handy with statistics, I thought I would see if there is a correlation between the liberal voting tendency of cities and several measures of the well-being of cities, starting with the unemployment rate. Most people realize that zero percent unemployment is neither possible nor good. A small rate of unemployment, perhaps something in the range of 4–6%, is indicative of a healthy labor market. With such an unemployment rate, employers and employees are reasonably able to find each other, matching jobs to the skills and preferences of workers.
But, with a much higher unemployment rate, many people seeking work will be frustrated in their efforts, and may even become discouraged from continuing to look. This is bad enough during times of recession. But, it is terrible for the unemployment to be permanently much higher than 4–6%. How can individuals find work, and provide for themselves and their families, when the unemployment rate is permanently high? How can individuals have the dignity that comes from work? And, how can we—as a society—hold individuals accountable for themselves, when it is not reasonably possible to find work?
Free-market economies—via Say's Law—tend to generate jobs sufficient to employ all those seeking work, although the process of matching jobs to job-seekers is sometimes difficult. But it is possible for the government to so interfere with the economy—through taxes and regulation, through subsidies to not-working, through tilting the tort process in favor of certain classes of litigants, and through failing to suppress crime—to diminish the ability of the free-market to provide economic opportunities.
Nowadays, the Bureau of Labor Statistics estimates local unemployment rates regularly for the fifty largest cities (in addition to doing the same for the nation as a whole, for each state, and for the larger metropolitan areas). In 2004, Detroit—America's Most Liberal City—was also the country's #1 large city for unemployment. While the unemployment rate in the fifty largest cities averaged 6.5%, it was 14.1% in Detroit.
The chart below juxtaposes unemployment against liberal voting tendency in the country's largest cities. On the vertical axis, I have plotted the unemployment rate for 2004, and on the horizontal axis, I have plotted an index of the rankings for liberal voting tendency published by the Bay Area Center for Voting Research. In this index, 100 represents Detroit, and lower numbers represent decreasingly liberal cities.
In this chart, notice that in 2004 the unemployment rate fluctuated about 5% for America's Most Conservative Cities, and tended to fluctuate about 7.5% for America's Most Liberal Cities. To be sure, some of America's Most Liberal Cities had relatively low unemployment rates, but the tendency was for the unemployment rate to rise with the tendency to vote liberal.
In the next chart, I present a correlation of unemployment rates with liberal voting tendency using data from the 2000 census. By using the 2000 census, I am able to include all the cities in my sample, including the mid-sized cities (from 100,000 to something like 250,000) as well as the large cities (above something like 250,000).
Back in 2000, prior to the bursting of the dot-com bubble and prior to the 9/11 attack on our country, the unemployment rate was quite low. In the cities in my sample, it averaged 4.2%. In some cities, the unemployment rate was even lower than 2%. Conversely, in some cities, the unemployment rate was higher than 8%. To my eye, it looks as though, in 2000, the unemployment rate ranged from 2–6% in America's Most Conservative Cities, and ranged from 2–10% in America's Most Liberal Cities.
With as much variation as there is in the above chart, I thought I should conduct a formal analysis. Based on some regressions I report in the appendix to this essay, it can indeed be said that unemployment tends to rise with liberal voting tendency. This tendency is demonstrated in both a simple regression of the unemployment rate against the index of liberal voting tendency, and a multivariate regression in which the independent variables include the size of the population and a variable that indicates suburban cities (as opposed to central cities) in metropolitan areas, in addition to the index of liberal voting tendency.
Liberalism and Crime
The second measure I looked at was the FBI's crime index. The FBI crime index includes six categories of crime (murder and non-negligent manslaughter, forcible rape, robbery, aggravated assault, burglary, larceny—theft and motor vehicle theft), and is simply the number of these crimes that are reported to law enforcement. Nowadays, the FBI publishes this crime index annually for cities and a variety of other jurisdictions (except not for jurisdictions within Illinois since they don't report rape statistics in a way that is consistent with the FBI request for data). It should be noted that it is thought that the FBI crime index suffers from nonreporting.
Without the suppression of crime, not only are our lives and property put at risk, but our liberty is destroyed. We, who are law-abiding, are forced to lock ourselves up within our homes and places of work, and limit our associations to those whom we already know. We are not free to travel, or to express ourselves, or to act on our beliefs. Hobbes said that men prefer tyranny to anarchy, but what about the possibility of a strong but limited government, one that protects us from crime without itself becoming abusive of our rights? It seems that liberals who focus their attention on redistributing wealth lose sight of the first responsibility of government, which is to protect its citizens.
Detroit, America's Most Liberal City, does not particularly stand out in the FBI's crime index. In 2002, Detroit had a crime rate of 0.0884 per capita. While this is higher than the average crime rate in the cities in the sample (0.0597), it is not in the top ten percent of the sample. St. Louis, Missouri, with a crime rate of 0.1429 per capita is America's worst city when it comes to crime, and it's only at the 20th Most Liberal City in terms of voting tendency.
Below, I present the FBI crime index for the cities in the sample, for 2002, per capita, graphed against liberal voting tendency. With all the variation in the data, I'm not sure that Lee Majors—The Six Million Dollar Man—could tell if there is a correlation between the crime rate and liberal voting tendency using the eye-ball method of statistical analysis. The multivariate regression reported in the appendix indicates that liberal voting tendency is modestly correlated with the crime rate. (The more significant correlation exhibited in the simple regression must be disregarded, since it is apparently due to "excluded variable bias.")
As I mentioned above, the FBI crime index is thought to be effected by nonreporting. It is possible that the crime rate in Detroit and other liberal cities is understated because people often do not report crimes against property, thinking that little will come of the report. Perhaps the police tend to be under-staffed in liberal cities, or the people do not cooperate with the police in these cities, in the apprehension and conviction of criminals. An alternate explanation for the "low" crime rate in Detroit and other liberal cities is that, as after the third revolution in Liberia, there's nothing left to steal.
To see if, possibly, the crime rate of Detroit and other cities is understated because of the nonreporting of property crimes, I thought to focus on murder and non-negligent manslaughter. I figure that even if the police are not very effective in the pursuit of justice when it comes to property crimes, people will still report murder and manslaughter—how else would they get rid of the dead bodies?
Gary, Indiana, the 2nd Most Liberal City in America, turns out to be our murder capital, with a murder rate of 58 per 100,000 people. The second highest murder rate, and highest among large cities, 53 per 100,000 people, belongs to New Orleans, Louisiana, the 26th Most Liberal City. Number three is Washington, D.C., the 4th Most Liberal City, with 47. And, number four is . . . my, my . . . Detroit, with 41. It turns out that Detroit is very much like Washington, D.C., where former Mayor Barry once said, "If you take out the killings, Washington actually has a very, very low crime rate."
In the regressions reported in the appendix, liberal voting tendency is shown to have a very strong correlation with the number of murders and non-negligent manslaughters. The strong correlation between liberal voting tendency and the murder rate and only weak correlation between liberal voting tendency and overall crime rate warrant further discussion. While my first inclination is to attribute the difference to the under-reporting of property crime in liberal cities, it must be considered that the difference has some validity. As to why the murder rate would tend to be higher in liberal cities, perhaps it is that liberal cities have stricter gun control laws, denying their residents' right of self-defense, and perhaps it is due to the deterrent effect of the death penalty in conservative cities. It might be worth investigating some of these possibilities.
Thus far, we have found that cities with more liberal voting tendencies tend to have higher unemployment rates and higher murder rates, and weakly tend to have higher overall crime rates. For a measure of the bottom-line, total impact of liberalism on cities, I thought I would look at the correlation between liberal voting tendency and population changes. Cities that are growing faster than the average are places that people feel improve their happiness by more than the cost of relocation. Considering that it is, by far, our country's largest city, the growth of New York City, from 1990 to 2000, by 9.4%, or almost one million people, is truly phenomenal. Detroit, the poster child of liberalism, lost 7.5% of its population during the 1990s.
It was not so long ago that New York City was falling into the liberal sink hole of deficit spending, increasing taxes, middle-class flight, rising crime, demoralized police, and collapsing infrastructure. Then came Rudy Guiliani, and, with him, a degree of spending restraint and some tax reductions (both of which remain very high), and a return to law and order. For most Americans, Mayor Guiliani is one of the real heroes of 9-11, but for New Yorkers, he is also responsible for rescuing the city from liberalism. In fact, New York City tends very strongly to vote liberal, being the 21st Most Liberal City. There is, therefore, no guarantee that this city will not return to its former ways when New Yorkers feel that it is OK to be liberal again.
Looking at the chart below, of population change and liberal voting tendency, the scatter of points appears to telescope down. On the conservative side of the graph, some cities are shown to have hardly grown at all, while others are shown to be growing very fast (if I hadn't cut-off the vertical scale at 120%, the range of variation would appear to be even greater). On the liberal side of the graph, the range of variation collapses, and most cities are shown to be shrinking. The regressions reported in the appendix confirm what is apparent to the eye; namely, that liberal cities tend to be shrinking cities.
To sum up our findings, liberal cities tend to have higher unemployment, more murders and shrinking populations, and weakly tend to have higher overall crime (at least, as reported). We might think that, with governments that are accountable to the people, policies that tend to lead to such consequences would not be long supported by the people. As Abraham Lincoln put it, "you can't fool all the people all the time." Indeed, this is why Ludwig von Mises endorsed democracy. Classical liberals support democratic governments not because the majority is right, but because majorities change.
But, this supposition does not apply to modern American cities. Today, dysfunctional cities can rely on transfer payments from other jurisdictions to sustain them. Their governments receive much of their funds from state and federal government through intergovernmental transfers, and many of their people, likewise, directly receive money and in-kind transfers from government authorities that can tax people outside their city's jurisdictional borders. Indeed, politicians within dysfunctional cities learn to cultivate dependency and class envy among the population that remains.
|Before Adam Smith: $25|
If these dysfunctional cities were independent countries, like North Korea or Zimbabwe, they would proceed rapidly to hyperinflation, anarchy or dictatorship. But, because they are sustained by transfers from healthy economies outside their jurisdictional borders, they can continue in disarray. Eventually, the state and federal governments that sustain these cities have to weigh the rights of the people within these cities, and especially the children who are wards of their "village," against the exercise of self-government.
As of today, we have only seen a handful of cities lose, or partially lose their power of self-government, through state and federal government takeover of their schools and through financial control boards. As voting becomes more and more polarized, and liberal places becoming increasingly committed to redistributing wealth instead of to the creation of wealth, the suspension or partial suspension of self-government in dysfunctional cities will likely develop into a trend.