City Journal

Stephen G. Craig
Where the Money Goes
A study of large American cities reveals that New York spends almost twice as much as average. Here's why
Autumn 1991

Does New York City spend too much money? And if it does, why are basic city services so often inadequate? These seem like obvious questions to ask as the city struggles with its second major fiscal crisis in 15 years. They seem especially relevant since the city leadership seems more willing to raise taxes than to cut spending, and the budget cuts that are being made seem to come mostly out of basic services of the sort that any moderately prosperous city ought to be able to afford.

Oddly enough, these questions do not get asked very often or answered very clearly. Part of the reason may be that they are hard questions to answer, at least with any precision.

One could say that since New York is going broke, it must be spending too much. But a budget crisis does not necessarily mean the city is spending too much: It could be taxing too little, though New Yorkers will be forgiven for being skeptical of this explanation. Besides, going broke is a pretty painful method for detecting fiscal imprudence.

One might try to establish how much money the city needs to spend. Good idea, but extremely difficult to pull off. Try defining need. Does New York need to eliminate crime? drugs? poverty? Obviously the city cannot do away with any of these things; it must choose between doing more or less to combat any one of them.

Probably the most useful way to measure whether we spend “too much” is to compare how much New York spends, and on what, with other similar American cities. Unfortunately this is hard to do: There are no standardized accounting procedures for American cities, and the services that are provided within a city’s borders may be offered by a number of different jurisdictions, which may have boundaries different from the city’s.

New York City, for instance, runs its own school system, but in many places school districts have boundaries and budgets quite separate from the city’s. On first glance New York might seem to spend a lot more on education than, say, Chicago, but raw comparison doesn’t mean much.

This is why economists and budget experts hardly ever compare spending across city lines. With sufficient time and care, however, it can be done. I performed just such an analysis, comparing New York City’s expenditures with those of ten other major American cities.

The state of the data and the inherent difficulties of the task required some compromises along the way. But fortunately for the clarity of the study, though perhaps unfortunately for New York, the results were so dramatic as to override any doubts about the technical details.

In brief, New York City is in a class by itself in its level of municipal spending. Even after controlling for population, differences in jurisdictions, and similar complications, no other city studied even came close. For every man, woman, and child living in New York, the city government (which makes essentially all local government expenditures here) spent roughly $3,000 in 1986, the last year for which comparable data for the various cities are available.

As Figure 1 shows, New York spent far more than the average city, and well above the figure for Los Angeles, the second highest-spending city on the list. Local government spending came to just under $2,000 per capita for all jurisdictions within Los Angeles County. (The City of Los Angeles by itself only spent about $650 per citizen.)

On average, New York City spends 75 percent more per capita than the local governments of the ten other cities. (If only central city governments are taken into account, New York spends well more than three times the average.) New York City spends more than twice as much per citizen as Boston and Jacksonville, and almost twice as much as Dallas and Chicago. (The cities studied were Baltimore, Boston, Chicago, Cleveland, Dallas, Detroit, Houston, Jacksonville, Los Angeles, and Philadelphia.)

To put it another way, if New York’s per capita spending were no more than average for typical large American cities, it would have run a $9 billion surplus in 1986. Today the figure would be even higher.

What accounts for the enormous difference between New York and these other cities? Not New York’s higher cost of living. If higher prices were the cause, then New York’s higher spending would be more or less evenly distributed among various government services. But that is not the case.

Indeed, one of the most surprising results of this study is that New York’s spending for traditional, basic municipal services is not particularly high. New York’s spending is just slightly above average for fire protection, parks, education, and central administration. We spend more for police than do the four “modern” cities on the list (Los Angeles, Jacksonville, Dallas, and Houston), but not much more than the six other “traditional” industrial cities of the Northeast and Midwest. As a percentage of its total budget, New York spends far less than other cities on these basic services.

New York’s per capita education spending, for example, is almost exactly equal to the average of the traditional cities and is actually 2.5 percent below the average of the four modern cities. Spending per capita is not the same as spending per student, and spending a lot per student does not necessarily ensure good education. Even so, it is clear that New York devotes less of its resources to its schools than other cities: 22 percent of its budget, compared with 39 percent for the average of the other ten cities.

Were it possible to accurately figure into this comparison the difference in the prices of goods and services among the several cities, the results would probably show New York to be underspending on basic services compared with the other cities. Yet New York’s leaders are preoccupied with balancing the budget and seem quite willing to do so at the expense of essential services.

So where does the money go? Again the answer is so clear and dramatic that none of the technical difficulties of cross-city comparisons can hide it. New York’s excessive spending is due almost entirely to the simple fact that, alone among the cities studied, New York City has undertaken to run its own welfare state.

Almost everywhere in this country the care of the poor, or at least such care that requires any large expenditure, is primarily a state and federal matter. But relative to other cities, New York City spends enormous sums on low-income assistance programs. Such spending is reflected primarily in four areas: welfare (including Medicaid), municipal hospitals (most of whose patients are poor), other health expenditures, and housing programs. New York spends $1,084 per capita on these four categories some 36 percent of total city expenditures. The average for the traditional cities is $252, or 14 percent of total spending; the average for the modern cities is $230, also 14 percent of total spending. (See Figure 2.)

Of course, much of this money never makes it into the pockets of poor New Yorkers. These figures include all administrative costs. Moreover, spending for health, hospitals, and housing goes entirely to service-providers, not directly to the poor.

For spending on welfare and hospitals, no other city even comes close to New York. New York spends 133 percent more on welfare than Los Angeles, the second highest-spending city, and 75 percent more on hospitals than Cleveland, the Big Apple’s closest rival in that category.

Looked at another way, the $839 per capita New York spends on hospitals and welfare is 28 percent more than the $658 per capita it spends on education. No other city spends even half as much on welfare and hospitals combined as on education.

Another way to illustrate the enormous proportion of New York’s budget that goes to low-income assistance is to subtract these four categories from the total spending figure. This leaves New York with an aggregate budget of $1,916 per capita, only 29 percent greater than the traditional-city average of $1,491. New York’s commitment to low-income assistance accounts for nearly two-thirds of the difference in expenditures between New York and other major metropolitan areas.

Of course part of the reason that New York City spends so much in the welfare category (including Medicaid) is state policy. Most other states cover a greater share of welfare costs than New York State does. But state policy does not tell the whole story: New York City spends more than three times as much per capita on welfare as other localities within New York State. New York City spends $562 per capita on welfare. The average for localities in the rest of the state is only $174, which is within the range of Cleveland and Los Angeles.

The conclusion is inescapable: New York City is short of money to pay for normal municipal services because it pays for things most cities do not. If we were not trying to run our own welfare state, whether at Albany’s insistence or our own, we would be able to spend far more on basic services and cut taxes in the bargain.

Two additional factors help make us a high-spending city: Our effective payments for debt service (including retirement spending) and our miscellaneous or “other” government expenditures are both much higher than in most other cities. In the main, though, it is New York City’s unique municipal welfare state that is causing our repeated fiscal crises and forcing us to cut basic services.

New York, with its great wealth and high tax rates, has far more resources at its disposal than any other U.S. city. Yet the average New Yorker does not enjoy the benefits of a wealthier local government, because the additional money is not spent on basic services that benefit all New Yorkers.

Tables 1-3 show, step by step, how I arrived at these conclusions. The 11 cities studied were selected to represent the most populous states, and they comprise half of the 22 largest cities in the country. The six traditional cities are shown separately from the four modern cities, and New York City is shown by itself. New York is more similar to the traditional cities, but the modern cities are worth examining as well, because increasingly they are our competitors for people and businesses.

As noted, comparing expenditures among cities is a perilous business, so conclusions must be drawn with care. There are no standardized accounting rules, so we must rely on census information, which is why I had to use 1986 data. Even with the census data, however, there are some minor inconsistencies. Nonetheless, the information is the best available.

The step-by-step analysis attempts to account for differences in spending caused by the different systems of local government and different divisions of responsibility between the states and their localities.

Intercity Comparisons

Table 1 shows only expenditures made by the central municipal governments of the 11 cities. This table shows that New York City spends a total of $3,064 per resident on its “general expenditure” budget—roughly 80 percent more than Boston, the next highest city, more than two-and-a-half times as much as the average traditional city, and about four-and-a-half times as much as the average modern city.

Because municipal responsibilities vary so wildly, the Table 1 general expenditure figures really do not help us very much in figuring out whether a particular city is spending “too much” overall, compared with other cities. They do, however, show that New York has entrusted a single central government with vast and complex responsibilities.

The more interesting figures in Table 1 are those for police, fire protection, and parks and recreation. These services are displayed in Table 1 because they are nearly always provided by municipal governments. (A notable exception is parks and recreation in Chicago.)

New York City’s spending on these basic services is close to the average for the traditional cities and somewhat higher than for the modern cities, especially for police. New York, for example, spends $176 on police, only 4 percent more than the traditional-city average, and $81 on fire protection, slightly above the average for traditional cities, whose expenditures range from $59 in Philadelphia to $107 in Boston. On parks and recreation, New York’s spending is slightly below the average, excluding that of Chicago.

But as a percentage of total spending, New York spends much less on basic services than the other cities. For every $100 New York City spends, approximately $10 goes to police, fire protection, and recreation. In contrast, for every $100 spent by the municipal governments in the traditional cities, nearly $25 goes to these functions.

Table 2 attempts to correct for the most serious and obvious difficulty in intercity spending comparisons: varying divisions of responsibility among different local government units, whose borders are not identical. In other words, Table 2 adjusts for the fact that in one area local welfare spending may be a county responsibility, while in another it is a city responsibility; or that county and city lines may not be identical; or even that school districts may stretch across city lines.

Table 2 handles these difficulties by comparing per capita spending by all local government units within the county (or counties) containing the major city and its immediate environs. Among these governmental units are counties, school districts, cities, townships, and special districts. Even under this procedure, New York spends 50 percent more per capita than the next highest-spending urban area, Los Angeles; 72 percent more than the average traditional city; and almost 80 percent more than the average modern city.

Table 2 also shows total local expenditures for a variety of individual services. Expenditures on education and roads continue the pattern of Table 1: New York does not spend very much on basic services that benefit all citizens. On education, it spends only $1 more than the average of the traditional cities and actually spends less than the modern cities. Again, given New York’s higher cost of living, it may be, in effect, spending significantly less on education than most cities.

Nor can New York’s budget woes be blamed on a bloated central bureaucracy: New York’s $96 per capita spent on central administration is equal to the traditional-city average. This category does not represent the total cost of bureaucracy; it includes such functions as finance, building maintenance, courts, and central executive offices, but not administrative costs for particular departments, which are included in the expenditure figures for the specific categories.

Table 2 shows clearly that New York busts its budget primarily on direct and indirect assistance to low-income New Yorkers. New York City spends more than ten times as much on welfare as the other cities studied. Welfare spending alone accounts for almost 40 percent of the difference between New York’s general expenditure level and that of the average city. The next biggest welfare spender, Los Angeles, spends less than half as much per capita as New York. The only other local area that spends more than $100 per capita on welfare is Cleveland.

New York is one of the few cities in the U.S. that owns any municipal hospitals. But even though Table 2 includes expenditures on county hospitals, which are somewhat more common, New York City spends almost four-and-a-half times as much on hospitals as the average traditional city, and nearly four times as much as the average of all ten cities.

According to Table 2, New York spends almost twice as much as the traditional cities spend on housing programs and more than four times as much as the modern cities spend.

One possible explanation for all this spending on the poor, of course, is that poverty is worse in New York than elsewhere. Trying to account for need, however, is a daunting task. If more New Yorkers are on welfare, is this because New Yorkers are poorer, or because it is easier to get welfare in New York? If New York’s poverty rate is no higher than elsewhere, does this mean there is no greater need for welfare, or does it mean New York’s generous welfare programs should be credited with keeping the poverty rate down?

In spite of the difficulty in determining whether New York is especially needy, it seems unlikely that need alone could account for the enormous difference between New York and other cities: New York’s total low-income expenditures are more than four times as great as the average traditional city. Moreover, if New York’s need for poverty programs is significantly higher than that of other cities, one would also expect New York to have greater needs in such areas as police and education. Yet, as we have seen, its spending on basic services is not particularly high, in part because it spends so much on welfare, housing, and hospitals.

The other major area in which New York significantly outspends other cities is debt service. New York spends $119 per capita for interest payments, some 20 percent more than the traditional-city average of $99 per person. If fiscally strapped Philadelphia is excluded, New York’s interest payments are 55 percent higher than the traditional-city average. New York’s interest payments are even slightly higher than the modern-city average, though newer cities would be expected to incur greater debt as they build new facilities to meet the demands of rapidly growing populations.

Moreover, New York’s interest-spending figure excludes a huge hidden debt in the form of increased retirement expenditures undertaken to compensate for the wage deferrals of the 1970s. The city, in effect, borrowed money from its employees during the last fiscal crisis and is now paying it back in the form of bigger pensions. Thus New York spends more than three- and-a-half times as much as the average traditional city on retirement expenses and more than six times as much as the modern cities. (This category is not included in the general expenditure total.) New York’s retirement expenditures are more than twice as high as those of the next highest city, Boston. A significant portion of these extraordinarily high pension payments should really be considered debt service.

The city’s huge debt-service obligations are one important reason that the city has so little fiscal maneuverability during hard times: Debt payments are for the most part intractable legal obligations that must be met, even at the cost of cutting basic services.

New York is also higher in the “other” category, which comprises all expenditures not otherwise classified in the census data. New York spends $269 in this area, compared with $110 for the average traditional city and a mere $48 for the average modern city. Since these services are not specified, the only conclusion one can draw is that New York offers a range of minor government services far beyond that offered by the traditional big city. (Some categories used by the Census Bureau are omitted from this analysis; these are not included in the “other” category, although they are part of the general expenditure total. See table notes for a specific definition of each category.)

The Role of State Policy

So, why doesn’t New York stop its free-spending ways, or shift some of its resources to the sort of basic services that might help keep its middle-class taxpayers from fleeing the city and making its fiscal problems worse? One often-cited reason is that Albany forces the city to spend a lot of money, particularly on the poor. A number of budget analysts and prestigious policy organizations have called on Albany to reduce the burden of state-mandated local spending, and in particular to pick up the full state share of welfare and Medicaid instead of shunting half of it off onto local governments.

But Albany’s responsibility for the city’s plight may be exaggerated. Table 3 attempts to determine to what extent city spending here and elsewhere is the result of state policies. It compares the rate of expenditure in the given metropolitan area with expenditures for local governments elsewhere in the state.

The numbers listed in Table 3 are the ratio of the Table 2 figures (local area per capita spending) to the corresponding figures for all other parts of the state. Generally speaking, the closer this number is to 1, the more likely it is that the local spending level is determined by state policy. A divergence from 1 indicates that the priorities of the city differ from those of other regions in the state. For New York City, the Table 3 ratio for general expenditures is 1.47; thus the city spends 47 percent more per capita than other localities in the state, though all are subject to the same state mandates, including those requiring them to share the cost of welfare and Medicaid.

For most of the individual categories, New York does not vary much from the other ten cities. For instance, New York spends 1.34 times as much as other New York localities on central administration. That is not much different from the 1.4 and 1.13 figures for the traditional and modern cities, respectively.

In five categories, however, the difference is dramatic. The first is education: New York spends only 69 percent of what other New York localities spend. The only other city that spends a similarly small percentage on education is Baltimore, at 76 percent. Seven of the remaining nine cities spend more than the state average.

The other four categories are, predictably, welfare, housing, hospitals, and retirement expense. New York spends more than three times as much on welfare, nearly five times as much on hospitals, and more than seven times as much on housing as other New York localities. The other cities, on average, had spending levels much more typical of the other localities in their states. New York is the only major locality in the state with significant retirement expenditures, so its Table 3 figure is particularly large.

The effect of state policy on New York City is most pronounced in the welfare category. Thus, while New York spends more than 12 times as much per capita on welfare as the average traditional city, it only spends about three times as much as other localities in New York State. Still, despite the heavy influence of Albany, New York City spends far more on welfare than other cities in the state, suggesting that New York City, on its own, could do a good deal to reduce its welfare spending.

Albany’s policies, on the other hand, do not seem to account for the difference between New York City and other New York localities in spending on hospitals and housing. In fact, New York City spends 492 percent of the state average on hospitals and 734 percent on housing. Both figures are more than the difference between New York and the traditional-city average, as shown in Table 2.

If New York followed the pattern of the average traditional city—spending 89 percent of the state average on welfare, 168 percent on hospitals, and 363 percent on housing—its total expenditures on these three categories would be $347 per capita, $689 less than the actual level, and only $140 more than the average traditional city.

A Thought Experiment

What would happen if New York’s spending patterns were similar to those of other cities? Two possible answers are given in Table 4, which shows two hypothetical budgets for New York City. These are not realistic budget proposals, of course. The city’s debt, for example, cannot be magically reduced. The hypothetical budgets are thought experiments designed to illustrate what might happen if the city’s priorities were more like those of other cities.

The first budget shows what would happen if New York simply spent the same amount per capita as the average traditional city in each of the categories. The result, in 1986, would have been a reduction in expenditures of $1,310 per capita, or a “surplus” of more than $9 billion for the whole city. A budget along these lines, however, cuts basic services such as fire and police. It also presupposes higher state expenditures on welfare programs, so a significant portion of this surplus would end up being paid in state taxes. Still, even if city welfare spending were kept at its current level, the surplus would amount to $793 per capita, or roughly $6 billion overall— enough to improve basic services and provide substantial tax relief.

The second hypothetical budget shows what New York’s spending might look like if it were more like those of other cities, but constrained by the state policies that now exist. The figures here are the amount New York City would spend if its per capita expenditure were the same proportion of the state average as is the case in the average traditional city. For example, Table 3 shows that the average traditional city spends 110 percent of its state average on education. If New York City spent 110 percent of the state average, its per capita education spending would be $1,049.

This budget produces only a $425 per capita surplus, about $3 billion citywide. But it considerably increases spending on basic services such as education and police. Many of the cuts are at the expense of programs for the poor, but a $3 billion surplus would allow considerable leeway to expand those programs if necessary. Debt service, as represented by interest and retirement expense, would be much reduced.

A central purpose of examining the city’s budgetary priorities is to expose what economists term the “opportunity costs”—that is, the cost of not spending money elsewhere. While these hypothetical budgets are not realistic proposals, they do show that New York spends relatively little on education, police, fire, parks, and roads.

These budgets show just how much New Yorkers sacrifice for our generous programs for the poor and for the fiscal irresponsibility of the past. Our leaders must ask or be asked—whether the city’s long-term prospects would be improved if the city gave a higher priority to the essential services that improve the quality of life for all New Yorkers.

SHARE
respondrespondTEXT SIZE
If you enjoyed
this article,
why not subscribe
to City Journal? subscribe Get the Free App on iTunes Or sign up for free online updates: