tax abatements. Concentrations of tax-exempt property are another problem in such cities as Boston and Chelsea.

A breakdown of the tax base by type of property is not available, but its composition is reflected in the other economic series which also indicate the nature of the local economic base.

Personal income is in a sense the ultimate measure of personal ability to pay as well as a crucial determinant of the demand for public services and an inverse index of certain needs. Although the variation is less than that in any other measure of the base, it is still relatively great compared to that in other metropolitan areas. It is important to consider personal income as it has a direct bearing on our evaluation of the fiscal situation in different communities. For example, Lexington which has a high median family income and a relatively low property base has less of a problem than Wilmington which has about the same property base but a far, lower income; and Everett with a large industrial tax base but a low income has less difficulty than Norfolk which has the same income but less than half the property base. The worst problems, of course, are faced by those cities which are low by both measures, such as Boston, Chelsea and Somerville.

The various measures of the economic base need not be correlated with one another, i.e., a town which is wealthy by one measure may be poor by another. This is so both because towns may specialize in a particular type of activity and because wealth of one kind, e.g., commercial property, may be separated from related economic activities, e.g., the income and residential property of the commercial managers. The various measures of specific activity vary greatly within the area, depending on the specialty of a particular community. We find wholesaling at strategic points on the periphery, such as Needham, Dedham, and Natick and in the central city. Manufacturing is focused in industrial centers along Route 128 like Waltham and Framingham. The service industry is located in Boston and a few other cities. Retailing is the most dispersed, depending on local purchasing power as well as on the existence of shopping centers, such as Peabody, Saugus and Framingham. Looking at Boston itself, we find that it is still a manufacturing, wholesaling, and service center, but that it has lost ground in manufacturing and retailing. Despite recent improvements, it still ranks near the bottom in taxable property value, in part perhaps because of the presence of much untaxable land. As noted earlier, more than 60 percent of all real property in Boston is tax exempt.

Local Government Expenditures

General expenditures per capita in the Boston area lie about the national metropolitan average, the Boston area's previous lead having been narrowed by its less rapid growth (Table 13). Spending in Boston central city is far above that for the rest of the area, but substantial differences exist among the cities and towns (Tables 14 and 15). Two fundamentally different sets of reasons give rise to two different groups of governmental expenditure patterns. On the one hand, we find the central city and similar older, poorer communities which are confronted with massive welfare problems, exploding protection demands, and the need for renewal. On the other hand, the new smaller towns, usually at least moderately wealthy, are confronted with massive capital needs as well as demand for high quality public services, especially education. The central city-suburb dichotomy is merely one view of the contrasting patterns which exist in the metropolitan area. At the one extreme, the City of Boston spends far more per capita on welfare programs, public health and hospitals, police protection, parks and recreation, general government, and special functions, such as urban renewal. The suburbs spend absolutely more on education, highways, and sewerage--all having large components of capital outlay.

If we look in detail at the variations in spending patterns among the individual towns, we see that total outlays per capita vary far less than do those for any particular function, reflecting the stage of community growth or maturity, the diversity

of purposes for which the governments spend money, and the general revenue constraint which forces them to trade off among potential uses of funds, in line with their stage of growth.

Capital expenditures in general are highest in the rapidly growing areas, tending to lag somewhat behind the growth in population. The largest expenditures are for new schools, with highways and sewer construction making up the bulk of the remainder. Replacement costs in the stable or shrinking older communities tend to be deferred as long as possible because of more pressing current needs and limited funds, with occasional spurts to catch up. This is most visible in the antiquated school buildings of Boston and some of the surrounding old cities. The least variable of the important functions is, of course, education, for which the basic need is well defined and many of the basic standards are uniform. The burden on a town depends not only on the age structure of the population, but also on the degree to which pupils attend private or parochial schools. Education outlays are lowest in the central part of the SMSA and in some of the older outlying cities, and highest in the wealthy suburbs.

The collection of welfare programs constitute the next largest category of spending, and, in the case of Boston and some other hard-pressed cities, the largest one, ahead even of education. The obvious explanation for the wide variations in spending among communities is the concentration of the various recipient groups in particular parts of the metropolitan area due to economic and social pressures. Casual observation suggests, however, that variations in the performance of the welfare function can be substantial, even given the framework of state standards and supervision.

Highway spending also imposes a widely varying burden; in part this view is amplified by the uneven nature of capital outlays from one year to the next. This is reflected in the fact that the highest per capita expenditures are in isolated outlying suburbs, such as Sudbury, Sherborn, Topsfield, and Manchester, which are rapidly opening up to development. The level of spending generally decreases as we move toward the center, with the established, moderately older communities spending less, and hits bottom in Boston. Expenditures per mile of road obviously move in a reverse fashion, because the higher population density toward the center permits substantial economies on a per capita basis. Nonetheless, we cannot help observing the sorry state of most of the central area highway system.

Given the increasing public concern with the protection functions, particularly police, it is interesting to note the wide variations in spending for these purposes. Per capita outlays are highly correlated with population density, as a proxy for all the conditions which lead to a demand for increased protection. The exceptions are the coastal resort communities, which spend relatively large amounts given their modest populations. It should be noted that in the Boston SMSA, and in the central city particularly, spending for both types of protection is very high in comparison with similar communities elsewhere. The City of Boston spends 50 percent more per capita on fire protection than does New York City; at the other extreme, some of the towns spend virtually nothing for this purpose.

Recreational and park facilities receive most attention in the larger wealthy suburbs and in the City of Boston, with most other communities spending very little. Public hospitals exist in only 7 of the 76 towns and cities, with the remainder having tiny public health programs if anything. The hospital cities carry the burden for the whole area, in general being reimbursed for far less than the costs they incur in providing services. The vast majority of the SMSA's renewal and public housing activities are in the city of Boston, which for its size has one of the largest programs in the country.

Sources of Local Revenue

The Boston SMSA leans more heavily on the property levy to finance local government outlays than do similar metropolitan areas across the United States, with the state providing a somewhat less than average share of general funds (Table 16). Much less use is made of user charges and fees or other types of taxes at the local level.

General revenues per capita in the city of Boston are almost 75 percent above those for the rest of the metropolitan area (Table 17). The City receives a larger share of its funds from both the state and federal governments than do the surrounding communities (Table 18). It relies even less on the use of charges and fees, and receives a smaller portion of its funds from the property tax. The debt picture is virtually identical, although individual communities in the outlying area have widely different debt burdens.

Looking at the detailed data for 1962, we note far greater variations among the towns of the metropolitan area as to the level of general revenues and the extent of outside financing. The use of nonproperty tax revenue sources displays a distinct discontinuity. The majority of towns get less than 7 percent of their local funds from such sources, but a small group receives over twenty percent. They are mostly outlying suburbs, such as Lynnfield, Burlington, Randolph, and Pembroke, which depend on fees for services and special assessments for some capital improvements.

Fiscal Effort

Local revenue per capita is a crude measure of the fund-raising task undertaken by local governments, irrespective of their underlying financial ability to raise revenue. This measure varies in a fashion similar to that of general expenditures per capita, although it is far from perfectly correlated, as the role of outside funding differs among the towns.

In measuring effort, we must choose some index of ability against which to measure performance. The two we choose are local revenue as a percent of personal income received by the residents and the effective property tax rate, corrected for differences in assessments (Table 19). These measures are partial and ignore the issues of tax incidence, both categorical and geographic.

Reliance on the property tax source in the Boston area is unusually heavy and, not surprisingly, it has some of the highest effective rates in the country. Rates cover a range of almost four to one: from Everett, an industrial enclave, to Chelsea, a poor old stagnant city with much tax-exempt property. Effective rates vary less than property tax collections per capita, implying that towns with extremely large and small tax bases have low and high rates, respectively. For example, at the area average tax rate of $34, Chelsea could have raised $96.50 per capita, Watertown $209, and Manchester $374; their actual rates, respectively, were: $75.50, $30.90, and $23.00.

The effect of a larger tax base is two-edged; it permits easier financing and may be associated with greater demand for high-quality services; on the other hand, it may be correlated with a lesser demand for other programs, such as welfare, public housing, and medical care. For the Boston area, a casual estimate would be that a town with twice the taxable capacity per capita of another would spend about 80 percent more, per capita (an expenditure elasticity of about 0.8). This means that the wealthier towns spend more on balance, other things being equal, but not in proportion to their wealth, so their effective tax rates tend to be lower. This fact is borne out by a fairly strong negative correlation between tax rates and taxable value per capita. The highest rates are to be found in the central city and in other old cities, such as Chelsea, Malden, and Somerville--all with true rates over $50 per $1,000. The lowest ones, in addition to Everett, are in the modest sized, fairly wealthy outlying suburbs like Duxbury, Manchester, and Wenham.