2

An Economic Framework for Assessing the Fiscal Impacts of Immigration

Thomas MaCurdy, Thomas Nechyba, and Jay Bhattacharya

INTRODUCTION

By 1994, almost 1 in 11 U.S. residents were foreign born. More than 9 million of these 23 million immigrants entered the United States in the past decade alone. As the immigration level rises, concerns are growing over the extra burden immigrants place on the government through their use of services such as welfare, health care, and schools. In reaction, states and the federal government have moved to limit the availability of services to both legal and illegal immigrants. The new welfare reform is the latest example. This legislation bars illegal immigrants from virtually all public benefits. It also bars or permits states to bar legal immigrants from major federal programs including cash welfare, food stamps, Medicaid, and Supplemental Security Income, although these provisions are currently under challenge.

Such anti-immigrant measures have been fueled in part by a growing number of studies that attempt to quantify the fiscal impact of immigration. Closely linked to literature on rates of immigrant participation in welfare and other social services, these studies try to carefully account for the taxes contributed and benefits received by immigrants at the federal, state, or local level. The exact items included vary widely. For example, most studies account for use of social services, health care, and schools, but others also consider services such as libraries, highways, community colleges, and parks. Similarly, most count immigrant payments to income taxes and often sales taxes, but others also include contributions to excise taxes, motor vehicle fees, even lottery revenues. Finally, the techniques for assigning the dollar amounts are quite different, such as assigning



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 13
The Immigration Debate: Studies on the Economic, Demographic, and Fiscal Effects of Immigration 2 An Economic Framework for Assessing the Fiscal Impacts of Immigration Thomas MaCurdy, Thomas Nechyba, and Jay Bhattacharya INTRODUCTION By 1994, almost 1 in 11 U.S. residents were foreign born. More than 9 million of these 23 million immigrants entered the United States in the past decade alone. As the immigration level rises, concerns are growing over the extra burden immigrants place on the government through their use of services such as welfare, health care, and schools. In reaction, states and the federal government have moved to limit the availability of services to both legal and illegal immigrants. The new welfare reform is the latest example. This legislation bars illegal immigrants from virtually all public benefits. It also bars or permits states to bar legal immigrants from major federal programs including cash welfare, food stamps, Medicaid, and Supplemental Security Income, although these provisions are currently under challenge. Such anti-immigrant measures have been fueled in part by a growing number of studies that attempt to quantify the fiscal impact of immigration. Closely linked to literature on rates of immigrant participation in welfare and other social services, these studies try to carefully account for the taxes contributed and benefits received by immigrants at the federal, state, or local level. The exact items included vary widely. For example, most studies account for use of social services, health care, and schools, but others also consider services such as libraries, highways, community colleges, and parks. Similarly, most count immigrant payments to income taxes and often sales taxes, but others also include contributions to excise taxes, motor vehicle fees, even lottery revenues. Finally, the techniques for assigning the dollar amounts are quite different, such as assigning

OCR for page 13
The Immigration Debate: Studies on the Economic, Demographic, and Fiscal Effects of Immigration average service use and average tax versus tracking actual use through administrative records. Not surprisingly, the resulting calculations reach quite different conclusions, some finding net costs of immigrants, others finding net benefits. In the absence of a systematic methodology, it is difficult to evaluate the competing claims. In this chapter we propose a basic economic framework for evaluating the fiscal impact of immigrants, addressing the complicated issues of demographic characteristics, skill levels, multiple levels of government, and the dynamic effects of a changing population. Although immigrants are the principal example for this framework, it has application beyond the immigration debate. In reality, immigration is merely population growth with particular population characteristics. The model seeks to separate these two aspects: expanding population and the change in population characteristics. The same framework could be used to model other population changes such as a baby boom. Immigration is a special case primarily because it can be influenced more easily by policy choices. In the second section we set out our basic economic framework for assessing the fiscal impacts of population growth. The third section begins with the specification of a ''neutrality proposition" that identifies an economic environment that makes population growth fiscally neutral. In that section we explore the factors that are relevant in evaluating the consequences of different patterns of growth, distinguishing between population increases arising from uniform shifts in all groups and disproportionate increases in particular groups, such as the elderly or the unskilled. In the fourth section we look at the particular nature of U.S. population growth due to immigration, applying the results of the third section to unveil the types of fiscal costs and benefits likely to accompany immigration. Finally, in the fifth section we review how the existing literature fits into our framework, permitting us to surmise the costs and benefits improperly assessed or missed in these various studies. ECONOMIC FRAMEWORK In this section we develop an elementary economic model for analyzing population growth in general and immigration in particular. Although our proposed model may at first appear more complex than necessary, we believe it is the simplest possible framework that can address the basic issues faced by researchers attempting to conduct fiscal impact studies of immigration. For reasons detailed below, we conclude that at a minimum the model must include the following features: multiple periods, three generations, workers distinguished by high and low skills, two consumption goods categorized by the level of sales tax,

OCR for page 13
The Immigration Debate: Studies on the Economic, Demographic, and Fiscal Effects of Immigration an underlying model of production and consumption, and a government sector detailed enough to capture the major categories of spending and taxation. In the following three subsections we provide the reasons for introducing these complexities and explain the role of these six features within the basic structure of the model. In the fourth subsection we explore some special concerns in using this general framework for fiscal impact studies of immigration relating fiscal trade-offs across periods and different levels of government. Multiple Generations and the Composition of the Work Force The first three elements of the model involve the composition of the population across time. There are three generations of agents living simultaneously in each period: the young (J), the middle aged (M), and the elderly (E). The work force is composed exclusively of the middle aged, and they are divided into high-skilled (h) and low-skilled ( ) workers. In conjunction with capital (K), the workers in the M generation produce current income (through the process described in the following subsection). Generation M consumes part of its labor earning and saves the remainder for consumption in the next period. The middle aged also provide for the young generation's consumption. The elderly consume from their holdings of capital (and government transfers). After every period, each generation advances, with E exiting the model and a new generation of J entering the model. Income, Production, and Consumption Aggregate income (Y) is produced through a production function ƒ which takes the two types of labor as well as capital as its arguments; that is, Throughout most of this chapter, we assume that this production function is constant returns to scale. Each type of M (high skilled and low skilled) is endowed with one unit of labor that is sold at the market wage. Wage rates are equal to the marginal product of labor, and the sum of wages is denoted by W. Similarly, the rental rate r is equal to the marginal product of capital. Income from capital is thus rK. Total consumption (C) by agents in the model is divided into two categories: Co (which is taxed at the normal sales tax rate) and Ce (which is subject to an additional excise tax), so that C=Co+Ce. Generation J does not earn income; it is sustained by generation M. The remain-

OCR for page 13
The Immigration Debate: Studies on the Economic, Demographic, and Fiscal Effects of Immigration der of generation M's after-tax income is savings invested to provide for income in the next period. Generation E's consumption is funded by reductions in its capital holdings. Let total tax revenues be equal to T (which is divided into its components in the following subsection). By definition, Y = W + rK = C + ΔK + T. Government Sector We now more fully incorporate the government sector into the model. To start, we treat this sector as one unit. In the fourth subsection, we consider the additional complications arising from the federal structure particular to the United States. The government sector taxes the various aspects of economic activity (consumption, wages, rental income) and spends resources on the production of "public" goods and on direct transfer payments to individuals in the economy. Our model considers each of these areas of government activity—public goods spending (Gp), spending on transfer payments (Gx), and collection of tax revenues (Gt)—by focusing on the characteristics of these activities that are relevant for thinking about the fiscal impact of population growth and immigration on government finances and on current taxpayers. To complete the model and to ensure that the accounting identifies hold, we then add a fourth government activity, the management of public debt (G d), that includes both making current interest payments and issuing sufficient bonds to cover any shortfall in current revenues. The various aspects of the government sector are summarized in Table 2-1. Gp: Spending on Public Goods In studying the fiscal impact of population growth and immigration, two characteristics of government spending on public goods are particularly relevant. First, relatively few of the goods produced by the government sector are pure public goods, in the sense that the cost of providing the same level of the good is invariant to the size of the population. Broadly speaking, we can therefore divide government expenditures on public goods into two stylized categories: pure public goods (Pp) that we define as goods subject to little or no crowding (e.g., national defense) and impure public goods or public services (Ps) that we define as goods that are subject to a substantial amount of crowding (public safety, education). This distinction is important in the analysis of the fiscal impact of population growth (through immigration or other means), because population growth entails additional spending only on public services. It should be kept in mind that crowding in public services can be nonanonymous, in which case not only the total number but the characteristics of those consuming the service

OCR for page 13
The Immigration Debate: Studies on the Economic, Demographic, and Fiscal Effects of Immigration TABLE 2-1 Government Sector = (Gp, Gx, Gt, Gd) Government Activity Subcategory Examples Role in Fiscal Impact Study Gp - "Public" Goods Production PP - Pure Public Goods National Defense Not Relevant   PS - Public Services       PC - Public Cons. Waste Disposal, Corrections ΔPC from Increased Pop.   Pl - Public Invest. Education, Infrastructure ΔPl minus Future Benefit Note: Crowding may be non-anonymous, and population externalities may be a factor. Gx - Transfers XY - Income Based T. Welfare, Medicaid ΔXY (Single Period Analysis)   XNY - Age Based T. Social Security, Medicare ΔXNY (Multiple Period) Gt - Tax Revenues τwW - Wage Tax Payroll Tax, Pers. Income T. Assumptions about tax incidence are important for taxpayer perspective; effect on tax bases matters for govt finance perspective   τKrK - Capital Tax Pers. Income T., Prop. Tax     τCC - Consumption Tax General Sales Tax     τeCe - Excise Tax Tobacco, Gas, Alcohol, Housing   Gd - Debt Management rD - Interest Expense   Not Relevant

OCR for page 13
The Immigration Debate: Studies on the Economic, Demographic, and Fiscal Effects of Immigration matter. For example, some school children require more resources than others. Also, population externalities of new population groups may be negative or positive, causing previous levels of services to become more or less expensive for the original population. (In public schools, for example, peer effects are thought to play an important role in school outcomes. Thus, the introduction of a "bad" peer group requires additional resources to keep the quality of schooling the same for the original population.) Second, studies on the fiscal impact of immigration must recognize that some public services are investments from the government's and the native taxpayer's perspectives in the sense that current expenditures on these services increase future tax receipts from the immigrant populations. Based on the degree of investment involved, we subdivide the category of public services into public consumption services (PC) and public investment services (Pl). Spending on public education, for example, can be thought of as an input in the production process that provides skills to the current J generation which then has implications for the distribution of skill levels in the future M generation work force. A higher-skilled labor force in the future implies higher incomes and consumption levels as well as increased savings and capital accumulation, all of which are taxed by the government. Spending on education therefore entails current costs and future benefits for the government sector, which implies that the long-run net fiscal impact of this type of spending differs from its instantaneous short-run impact on current government budgets. Other types of public services, such as most municipal services, serve primarily as current consumption and contain little or no investment. Government spending on public goods (Gp) is therefore divided (on the basis of crowding) into spending on pure public goods (Pp) and spending on public services (Ps), and public services are further divided (on the basis of the degree of investment) into public consumption (PC) and investment (Pl) services. Although the degree and kind of crowding (and the resulting distinction between pure public goods and public services) have implications regarding which types of government expenditures to consider in calculating costs of immigration, the degree of investment in public service activity (and the resulting distinction between PC and Pl) have implications regarding how such costs must enter the analysis. Gx: Spending on Transfers Much of today's government activity, however, has little to do with providing public goods and services and merely involves transfers from one population group to another. Transfer payments can be in the form of cash payments (Social Security, welfare), in-kind programs (Medicare, Medicaid, food stamps), or implicit tax exemptions (home mortgage deductions). In general, they fall into one of two broad categories: income-based transfers (XY) and non-income-based

OCR for page 13
The Immigration Debate: Studies on the Economic, Demographic, and Fiscal Effects of Immigration transfers (XNY. Income-based transfers include general assistance programs for the poor (Medicaid, food stamps) but may also include transfers to middle-class taxpayers in the form of such tax expenditures as the home mortgage deduction. Non-income-based transfers, on the other hand, include mainly programs that transfer resources based on age (such as Social Security and Medicare). The distinction between these two categories of transfers will become particularly relevant when we discuss the difference between general population growth and selective population growth through immigration. Gt: Tax Revenues Our model includes four broad categories of taxes: taxes on wages (W), rents (rK), general sales or consumption (C), and consumption of specific goods (Ce). Each of these activities is taxed at a specific rate (τ W, τk, τs, and τe, respectively), yielding total tax revenues T = τwW + τkrK + τsC + τeCe. In the United States, most tax revenues come from personal income taxes, payroll taxes, corporate income taxes, general sales taxes, excise taxes, and property taxes, each of which can be viewed as a combination of the taxes introduced above. Personal income taxes are generally assumed to fall on W and K, payroll taxes on W, and corporate income taxes on K (i.e., on all capital, not just corporate capital). General sales taxes are borne by consumers in proportion to their total expenditures on C, whereas excise taxes are borne by those consumers who consume Ce. (See Pechman, 1985, for a detailed discussion of assumptions leading to these tax incidence implications.) Finally, the property tax is borne by K, C, and W (see Mieszkowski and Zodrow, 1983; Aaron, 1975; Hamilton, 1975). Gd: Closing the Government Sector with Public Debt Management For the accounting identities to hold, we also include interest payments on current debt (r&273B;D where D is the value of outstanding government debt) as one of the expenditures incurred by the government sector. Furthermore, deficit financing (ΔD) covers any shortfall between current government revenues and expenses. Thus, total government revenues are equal to (T+ΔD), whereas total government expenditures are equal to (PP + Pl + PC + XY + XNY + rD). By definition. T+ΔD = PP + Pl + PC + XY + XNY + rD. For simplicity, from now on we assume that the government budget is balanced (ΔD = 0) in each period, although we allow for the existence of debt (D>0) and interest payments (rD>0).

OCR for page 13
The Immigration Debate: Studies on the Economic, Demographic, and Fiscal Effects of Immigration The Role of Costs and Benefits in Fiscal Impact Studies of Population Growth To determine the net fiscal impact of immigration (or any type of population change) within the context of the model presented above, three important elements must be explicitly addressed: (1) the definition of "costs" and "benefits," (2) the role of multiple periods in the analysis, and (3) the perspectives of different levels of governments in a federal system. Some conclusions regarding the first two issues are summarized in the last column of Table 2-1; those relating to the federal nature of the U.S. government sector are summarized below in Table 2-5. Definition of Costs and Benefits Our task in this chapter is to describe the fiscal consequences of changing population sizes and characteristics. The first critical step in this endeavor involves explicitly defining both the benefits and the costs of population changes. Fiscal impact studies have implicitly used different combinations of benefit and cost definitions, resulting in impact estimates that differ both in size and in interpretation. Costs, for example, have been defined as either the increase in total government expenditures resulting from the population change or the cost of government goods and services consumed by the new population, whereas benefits have been defined as either the increase in total tax revenues resulting from the population change or the value of taxes paid by the new population group. If all government expenditures were pure transfers (XY and XNY), definitions (1) and (2) would be identical; that is, the cost of the transfers to the new population group would be exactly equal to the increase in the total government budget. Suppose, however, that the government also produces pure public goods (PP). Then a population increase does not increase costs as defined by (1) but it does entail costs as defined by (2). Thus, some fiscal impact studies count part of national defense spending, for example, as a cost of immigration, whereas others do not. Similarly, the first definition of benefits ignores tax incidence and is concerned only with the impact on overall tax revenues (and thus the impact on tax bases), whereas the second definition concerns itself with the amount of taxes actually paid by the new population. Thus, the latter definition of benefits must be explicit about tax incidence assumptions. Different combinations of these definitions not only lead to vastly different fiscal impact estimates, but they also view the problem from very different perspectives. Table 2-2 summarizes these perspectives. From a government finance

OCR for page 13
The Immigration Debate: Studies on the Economic, Demographic, and Fiscal Effects of Immigration TABLE 2-2 Summary of Perspectives   Revenues   Expenditures Definition(1') Definition(2') Definition (1) Government Finance Perspective Native Taxpayer Perspective Definition (2)   New Taxpayer Perspective perspective, for example, the benefit of a population change is simply the increase in tax revenues resulting from this change, whereas the cost is the increase in government expenditures. Thus, Net Benefit to Government = (1') − (1). On the other hand, from the native taxpayers' point of view, the net benefit of a population change is the difference between after-tax income with and without the population change (assuming government services for the original population remain constant). This involves both the change in before-tax incomes and the change in taxes paid by the native population, where the latter is a function of the increase in government expenses due to the new population minus the tax revenues paid by that population. In the case of immigration, for example, the net benefit to native taxpayers involves both the effect of immigration on native incomes and the additional tax burden (which may be negative) incurred by natives to finance the net fiscal drain (which may similarly be negative) of immigrants on the government. Thus, Net Benefit to Taxpayers = Δ(Native Before-Tax Incomes) + (2') − (1). Finally, the fiscal impact on the new population could also be considered. In the case of immigration, this involves comparing incomes of immigrants in the United States with the incomes the immigrants would have earned in their native countries. More precisely, the fiscal impact of immigration from this perspective is equal to the change in the immigrants' before-tax income resulting from immigration plus the net benefit they receive from government services; that is, Net Benefit to Immigrants = Δ(Immigrant Before-Tax Incomes) + (2) − (2').1 Because the stated purpose of most fiscal impact studies of immigration is to calculate the net fiscal impact of immigration on U.S. taxpayers or government finances, definition (1) is the appropriate definition of costs (as it appears in both the "Net Government Benefit" and the "Net Taxpayer Benefit" equations). This implies that pure public goods (PP) and interest on debt (rD) are not relevant to calculating the fiscal impact of immigration because the addition of further popu- 1   This assumes that the immigrants would be obtaining zero net benefit from the tax/expenditure system in their country of origin.

OCR for page 13
The Immigration Debate: Studies on the Economic, Demographic, and Fiscal Effects of Immigration lation groups does not add to those government expenditures. Therefore, out of total government expenditures (PP+Pl+PC+XY +XNY +rD) only public services (Pl+PC) and transfers (XY+XNY) represent legitimate categories of government expenses to be counted as costs of population growth (and immigration). Although the definition of costs is the same from either the taxpayer or the government finance perspective, the definition of benefits differs between the two. The first definition (1´) ignores tax incidence and focuses on total tax receipts, whereas the second definition (2´) ignores total tax receipts and focuses on tax incidence. Most fiscal studies of immigration attempt to ascertain the impact on taxpayers rather than on government finances, which implies that (2´) is usually the appropriate definition of benefits. Still, fiscal impact studies have also been commissioned by local and state governments that may be more interested in the final impact on balance sheets rather than the degree to which immigrants are actually paying for additional tax revenues. In the former case, assumptions about tax incidence matter, whereas in the latter case assumptions about behavioral changes leading to tax base changes matter. The Need Multi-Period Analysis It is inappropriate to view the fiscal impact of population changes and immigration in a static one-period model. The age of the new population group has important implications for its impact on the economy (and thus the income of the current population) as well as its net impact on government finances. The three age groups in our model (J, M, E) each play a different role in the economy. To calculate the benefits and costs of an increase in any one of these age groups, both the present and the future must be taken into account—unless the increase occurs among E (who have no future role in the economy) or there is an expectation that the increased population will ultimately leave the economy before it becomes part of E (by returning to their country of origin, for example). We consider the impact of population growth among the three age groups in some detail in the third section of this chapter. For now, we merely stress the necessity of a multi-period analysis on both the expenditure and the revenue sides, given populations of different ages. Table 2-3 divides major areas of government expenditure into the broad categories defined in our model, indicating the age groups the spending programs mainly target. It is clear from the table that public consumption services are rarely age specific, whereas public investment services can be age specific, particularly education programs. Similarly, the bulk of income-based transfers—in terms of budgetary outlays, Medicaid—are not age specific, whereas the non-income-based transfer programs generally are age specific. In principle then, most of P C and XY could be analyzed within single periods. These expenditures are consumed today with few implications for the future (except that those con-

OCR for page 13
The Immigration Debate: Studies on the Economic, Demographic, and Fiscal Effects of Immigration TABLE 2-3 Public Expenditures by All Levels of Government Government Activity Specific to J Specific to M Specific to E Not Age Specific PP - Pure Public Goods       National Defense Int. Relations PC - Public Consumption Services       Environment Govt. Administ. Judiciary, Corrections, Police, Fire, Municipal Services Pl - Public Investment Services Education Head Start Tuition Subsidies Job Training   Infrastructure XY - Income Based Transfers AFDC Child Nutrition (Head Start) (Tuition Subsidies) AFDC Job Training   SSI, Medicaid Food Stamps Housing Assistance XNY - Non-Income Based Transfers (Education) Unemployment Social Security Medicare Retirement Benefits   suming income-based transfers today may be more likely to also consume them tomorrow). Non-income-based transfers (XNY), on the other hand, generally occur only in one specific period during the lifetime of an agent. In the case of Social Security, for example, if the growth in the population occurs in M, there is no Social Security expense today, but there will be an additional expense tomorrow when today's M becomes tomorrow's E. If new members of the M generation remain in the country as they become elderly, their expected future Social Security benefits must clearly be considered. Similarly, public investment services (Pl involve expense today, but the government sector will receive additional tax revenues in the future as a direct result of the expense. Thus, spending on education for J, for example, costs today but produces higher incomes (and thus income tax revenues) in the future when today's J becomes tomorrow's M. The share of Pl to be counted as a cost of population growth in J must therefore be reduced by the expected (discounted) future payoff from this expense. Expenditures on both public investment services and non-income-based transfers thus require a multi-period perspective. The need for this multi-period perspective becomes even clearer when revenues and expenditures are considered simultaneously. On the revenue side, Table 2-4 divides U.S. taxes into the categories in our model based on the genera-

OCR for page 13
The Immigration Debate: Studies on the Economic, Demographic, and Fiscal Effects of Immigration likely differences between immigrants and natives in consumption when calculating sales taxes paid by immigrants. Diverse studies of both illegal and legal immigrants, several of which are reviewed in the final section of this chapter, all assume that immigrants consume in the same proportions as natives in the same income category. This assumption, although understandable given the lack of data, is untenable given cultural differences between natives and immigrants. For example, immigrants commonly send a significant part of their earnings back to family members in their country of origin. These remittance payments contribute to neither sales tax revenues nor to U.S. savings or investments. (They may however contribute to the wealth or human capital of future immigrants.) As already noted, calculations of immigrant contributions to both current and future tax revenues critically depend on a better understanding of these behavioral traits. Violations of Condition (3): Characteristics Impacting Expenditures and Taxes Our discussion in the third section suggest that the characteristics of the new population groups may cause the marginal costs of extending public services to be higher or lower than average expenditures for the native population. This subsection explores two characteristics of immigrant populations that may in fact contribute to such higher marginal costs (and have important implications elsewhere). We highlight the role of parental education and socioeconomic status in public education crowding, and we discuss the problems for researchers that have been raised by the location choices of immigrant groups. Parental Education, Socioeconomic Status, and Cultural Factors in Education Immigrants' skills and education levels play an important role in the provision of education for their children. The best predictor of educational outcomes for children is parental education and socioeconomic status. Thus, parental characteristics play a crucial role in the amount of crowding and externality effects experienced by public schools from increases in the student population. Separately, low parental education levels and low fluency in English among immigrant parents induce greater funds to be spent on bilingual and bicultural education.20 As revealed in Table 2-13, immigrant groups from different countries of origin have substantially different fluency rates and thus cause substantially different levels of crowding in public schools. Finally, cultural factors stressing such values as work and individual achievement differ greatly among immigrants 20   Although there is a debate regarding the effectiveness of various forms of bilingual education, some form of bilingual (and sometimes bicultural) education is required in many states by state supreme courts.

OCR for page 13
The Immigration Debate: Studies on the Economic, Demographic, and Fiscal Effects of Immigration from different countries (Sowell, 1994). Borjas (1992) finds empirical support in the National Longitudinal Survey of Youth for the notion that "ethnic capital" plays an important role in the labor market and education outcomes of immigrant children, separately from the effect of parental human capital mentioned above. The role of culture, socioeconomic status, and parental human capital in determining differential rates of crowding in public schools, as well as in the use of other public services, are clearly significant and should play a prominent part in any fiscal impact study. Location Choice of Immigrants Immigrants do not locate themselves uniformly across the United States. The concentration of immigrants in certain geographic sectors may give rise to important threshold effects that would not appear if population growth were uniform. Although immigrant children might be absorbed easily into public schools if they were spread uniformly across the United States, this is clearly not the case when immigrant groups concentrate in a select number of states and their school districts are faced with massive fixed expenditures to accommodate those students. Past waves of U.S. immigrants often stayed in the first port where they arrived. For example, Irish, Italian, Jewish, and Puerto Ricans concentrated in New York and other port cities of arrival in successive waves over the nineteenth and twentieth centuries (Sowell, 1981). Today, the top four immigrant recipient states—California, New York, Florida, and Texas—receive nearly 60 percent of all new immigrants. Table 2-21, based on INS data, shows the state of intended residence of the 1987 and 1994 arrival cohorts. During this seven-year period, the top five states' share of new immigrants declined slightly, from 67 to 64 percent. Once immigrants arrive in the United States, they tend to migrate across state lines more frequently than do natives of the same age and ethnicity (see Bartel, TABLE 2-21 Intended Location of 1987 and 1994 Immigrant Arrival Cohorts   % of Arrival Cohort State 1987 1994 California 26.79 25.92 New York 18.98 17.95 Florida 9.09 7.22 Texas 7.04 6.98 New Jersey 5.13 5.48 Illinois 4.32 5.27 Other States 28.65 31.18 SOURCE: U.S. Immigration and Naturalization Service (1987, 1996).

OCR for page 13
The Immigration Debate: Studies on the Economic, Demographic, and Fiscal Effects of Immigration 1989:Table 7). Bartel (1989) also finds that it is the more highly educated immigrants who are more likely to move away from cities with a high concentration of immigrants of the same nationality. Finally, she finds that immigrants are more likely than natives to live in standard metropolitan statistical areas (SMSAs), and that whereas immigrants may move, they tend to move to other SMSAs. As we noted above in several different contexts, residential choice and probability of migration are important in determining current and future impacts of immigration. Congestion and externalities in the use of public goods will also be more serious concerns where immigrants are concentrated. It is not accidental that it is the states and localities that absorb the highest volume of immigrants, and especially illegal immigrants, that have commissioned reports calculating the net fiscal impacts of immigration on their appropriate level of government. (In the next section, we briefly review the recent fiscal impact reports commissioned by Los Angeles and San Diego counties, California, and Texas.) The location of the new immigrants and the consequent congestion problems are dependent on many of the characteristics of immigrants that we discussed above, including education level and national origin mix, and are likely to change over time, further necessitating a dynamic framework when evaluating the fiscal impact of immigration. CONCLUSION In this section, we review a number of recent fiscal impact studies through the lens of the economic framework developed in this chapter. This is not intended to be a comprehensive examination of the literature; Rothman and Espenshade (1992) already offer an extensive survey of the cost-benefit studies of immigration. Instead, we use this section to illustrate some of the shortfalls of analyses done without a comprehensive theoretical structure. Table 2-22 presents a summary of seven recent studies of the fiscal impact of immigration. The four sections of the table compare and contrast the studies, based on the population considered, the revenue sources and expenditure categories counted and their final "bottom-line" estimate. Using very different accounting strategies, most of the studies reach the conclusion that the immigrant groups considered are a net fiscal drain on the government. These seven studies define their targeted immigrant populations based on differing legal definitions and geographical areas. Most include illegal immigrants; two consider solely this group. Others add in recent immigrants, amnesty immigrants (Immigration Reform and Control Act legalized), and the citizen children of illegal immigrants in some combination. One considers all legal immigrants. Because many of these studies were sponsored by agencies at various levels of government, the geographical areas run the gamut from counties (Los Angeles County and San Diego County) to states (California, Texas, New Jersey among others) to the entire United States. As we discussed above, the

OCR for page 13
The Immigration Debate: Studies on the Economic, Demographic, and Fiscal Effects of Immigration TABLE 2-22 Summary of Fiscal Impact Studies Study Stewart, et al. (1992) Parker & Rea (1993) Huddle (1994) Romero, Chang & Parker (1994) Garvey & Espenshade (1996) Clark, et al (1994) Huddle (1993) Population Illegal Immigrants Yes Yes Yes Yes Unspecified Yes Yes Legal Immigrants post-1980 amnesty aliens, children of illegals No Post-1970 immigrants amnesty aliens citizen children of illegals Yes No post-1970 immigrants & amnesty aliens Scope LA County SD County Texas California New Jersey CA,FL,TX,NY,IL,AZ,& NJ U.S. Population Size 2.3 Million 220 K 1.9 Million 1.7 Million 39 K Various7 19.3 Million Revenue Sources Fed. Income Tax Yes1 No Yes Yes No No Yes2 State Income Tax Yes1 Yes Yes Yes Yes Yes Yes2 Sales Taxes Yes1 Yes ? Yes Yes Yes Yes2 Property Taxes Yes1 No ? Yes Yes Yes Yes2 Excise Taxes Yes1 Yes4 Yes3 ? Yes No Yes2 Payroll Taxes Yes1 Yes No No ? No No Immigrant-owned business Taxes No No No No No No No Expenditures Medicaid & county health Yes Yes4 Yes Yes1 Yes1 Yes5 Yes Social Services & Welfare Yes Yes Yes Yes Yes1 No Yes Educational costs Yes1 Yes Yes6 Yes1 Yes1 Yes1 Yes1

OCR for page 13
The Immigration Debate: Studies on the Economic, Demographic, and Fiscal Effects of Immigration Public housing No No Yes No ? No Yes Costs of criminal courts Yes Yes Yes No ? No Yes Incarceration costs No Yes Yes Yes1 ? Yes Yes Local public goods - parks, etc. No No No Yes1 Yes No No Bilingual Education No No Yes No No No Yes Local ''environmental costs" No No No Yes ? No No Student Aid No No Yes No No No Yes Social Security No No No No No No No Medicare No No No No No No No Earned Income Tax Credit No No Yes No No No No Fiscal costs associated with native job displacement No Yes Yes No No No Yes Final Estimates Revenues (Billions) $0.1410 $0.06 $1.47 $0.74 See note 8 See note 9 $283 Costs (Billions) $0.9510 $0.30 $6.15 $3.4 See note 8 See note 9 $952 1 Estimates rely on comparisons with observably similar demographic groups. (Comparison group varies by study). 2 Extrapolated from LA County tax receipts, and multiplied by an arbitrary "adjustment factor." 3 Assumes immigrants and natives consume these services at the same rate. 4 Assumes immigrants and natives consume these services at the same rate for some of the services in the category. 5 Emergency services only. 6 Unclear/ difficult to ascertain from description of methodology in paper. 7 Several different population size estimates are calculated and compared in this study. They vary depending on assumptions and data sources. 8 Final results are broken down by demographic groups and level of government - no single figure is presented. Net fiscal impact per immigrant household at state level varies between -$1000 and -$3000. 9 Varies by state and different demographic/behavioral assumptions used. 10 The study reports revenues and costs for other levels of government as well.

OCR for page 13
The Immigration Debate: Studies on the Economic, Demographic, and Fiscal Effects of Immigration choice of government perspective will change the conclusions of the study; the characteristics of the specific new population considered will also change the outcomes. Our first condition for population growth to be fiscally neutral is that expenditures on pure public goods and on public debt be zero. If these expenditures are positive, we conclude that population growth has a positive fiscal impact, all else being equal. Of course, these expenditures are positive. The studies in Table 2-22 treat these "fixed" public expenditures in one of two ways. They either ignore fixed public expenditures altogether, implicitly assigning zero benefit from immigration for these categories and thus biasing downward the impact estimate. Or they assign the per capita expenditures on these items as costs rather than benefits, which leads to an even larger negative bias. None of the studies attempt to calculate the benefit derived from immigration in "diluting" the base of payment for these pure public goods, even for major categories of expenditures as the debt, national defense, current Social Security transfers, 21 science research, and so on. How these benefits ought to be dealt with depends on which perspective the researcher takes and the level of government considered. It is obviously legitimate for a study on the county-level fiscal impacts of immigration to exclude the "dilution" benefit from federal programs. Nevertheless, some attempt should be made to estimate to what extent the goods and services provided by the county are pure public goods and count the effects appropriately. Some of the confusion would be eliminated if studies used marginal costs rather than average costs. The studies uniformly prefer the former, perhaps in order to maintain tractability. Although they do not explicitly consider returns to scale (our second neutrality condition), these studies focus exclusively on the direct effects of immigration on government finances, ignoring the indirect effects, especially the impact of immigrants on the income of the original population. Huddle's two studies (1993, 1994) are the exception. He estimates the number of native workers "displaced" by immigrants from their jobs and then calculates how much this unemployment would negatively impact government finances.22 All of the studies, including Huddle's, ignore the possibility that immigration might expand the job opportunities for natives in the long run, perhaps by new business creation, increased specialization, or a demand effect. In effect, they ignore the possibility that increasing returns to scale might hold. Of course, the empirical question of whether there are positive scale economies present in the American economy that can be exploited with immigration, or negative scale economies that suffer with 21   Because U.S. residents must work 40 quarters to qualify for Social Security benefits, the current costs of Social Security are "fixed" from the perspective of newly arrived immigrants (see Simon, 1989). 22   Passel (1994) vehemently attacks Huddle for ignoring or misreading the literature on the effects of immigration on native workers.

OCR for page 13
The Immigration Debate: Studies on the Economic, Demographic, and Fiscal Effects of Immigration the increased population, is unresolved. Nevertheless, as our framework suggests, these effects are likely to be important in the calculation of fiscal impact. Our neutrality proposition is also violated if the new and old populations have differing consumption and savings patterns. These seven studies attempt to estimate the effect of differing consumption patterns, but none discuss savings behavior. Remittances are discussed as a loss of sales and excise tax revenue. Taxes on immigrant-owned businesses are completely ignored. Although these studies do try to distinguish immigrant and native consumption patterns, they are less careful about distinguishing between immigrant groups. Several of the studies on illegal immigration assume that immigrants consume their income in the same way as other, more easily observable, legal immigrant demographic groups. Parker and Rea (1993), however, conducted a small-scale survey of illegal immigrants in San Diego County that includes questions about their consumption patterns. Unfortunately, because the illegal immigrants are identified mainly through their workplace, the sample chosen by Parker and Rea is not likely to be representative. Without better data, future fiscal impact studies of illegal immigration are likely to find this problematic as well. All the studies, in at least some of their calculations, assume that immigrants and demographically similar elements of the native population share the same propensity to use government services. This is particularly true when calculating immigrants' share of the costs of public schools, where per pupil costs are generally assumed to be the same as for natives. Rothman and Espenshade (1992) find the same assumption in all of the fiscal impact studies that they review. The only exceptions are again Huddle's two studies, which include the costs of bilingual education. None attempt to estimate crowding externalities induced by large influxes of immigrants on the use of these public services. It seems reasonable to guess that including the correct marginal costs and allowing for crowding externalities would increase the estimate of the fiscal costs. However, this omission could bias the results in either direction and may depend on the precise immigrant group being considered (due to language, culture, and socioeconomic differences). Ultimately, a serious shortcoming of these studies is the static model used for calculation. All existing studies ignore some critical intertemporal effects that impinge on how certain taxes and transfers should be counted. As our framework demonstrates, the current and future period effects could have opposing impacts on the fiscal costs and benefits of immigration. Many of the elements discussed in our synthesis have a dynamic component that should not be ignored. For example, education expenditures on immigrant children are invariably counted as a cost in the accounting schemes of the various papers. However, they are also an investment designed to make the young generation more productive in the future. Thus, the extra education expenditures result in future higher fiscal inflows that should be counted in the analysis at an appropriate discount rate. Another prominent example is payroll taxes, such as FICA, which is used to finance Social

OCR for page 13
The Immigration Debate: Studies on the Economic, Demographic, and Fiscal Effects of Immigration Security, that are left completely out of most of the studies. If these taxes are viewed as forced savings for retirement, then their omission from the analysis is justified. Alternatively, if, as we argue, they are intergenerational transfers from the middle-aged generation to the elderly generation, then the effect of the new population on these fiscal revenues and expenditures should be accounted for in some manner. Finally, the focus on consumption ignores the role of savings and its potential impact on future revenues, especially if they help the economy to grow in the future. Clearly, there is much work left to be done to acquire accurate estimates of the fiscal impacts of immigration at any level of government. Many gaps exist in our basic knowledge of the economic behavior of immigrants, both legal and illegal. This chapter seeks to improve our understanding of the theoretical underpinnings of fiscal impact calculations. In doing so, it potentially provides a road map of the necessary future steps for empirical research. ACKNOWLEDGMENTS This project was undertaken at the request of the Commission on Behavioral and Social Sciences and Education of the National Research Council (NRC). We gratefully acknowledge support from National Institutes of Health (NIH) grant HD32055-02. Opinions stated in this document are those of the authors and do not necessarily represent any official position or policy. We benefited greatly from Margaret O'Brien-Strain who helped us extensively with both the ideas and the editing of this chapter. Selen Opcin also provided editing assistance. REFERENCES Aaron, Henry. 1975. Who Pays the Property Tax? Washington, D.C.: The Brookings Institution. Bartel, Ann P. 1989. "Where Do the New U.S. Immigrants Live?" Journal of Labor Economics 7(4):371–391. Betts, Julian. 1995. "Does School Quality Matter? Evidence from the National Longitudinal Survey of Youth." Review of Economics and Statistics 77:231–250. Blau, Francine D. 1984. "The Use of Transfer Payments by Immigrants." Industrial and Labor Relations Review 37(2):222–239. Blau, Francine D. 1986. "Immigration and the U.S. Taxpayer." Pp. 89–110 in 1986 Essays on Legal and Illegal Immigration, S. Pozo, ed. Kalamazoo, Mich.: W.E. Upjohn Institute for Employment Research. Borjas, George J. 1990. Friends or Strangers: The Impact of Immigrants on the U.S. Economy. New York: Basic Books.

OCR for page 13
The Immigration Debate: Studies on the Economic, Demographic, and Fiscal Effects of Immigration Borjas, George J. 1992. "Ethnic Capital and Intergenerational Mobility." Quarterly Journal of Economics 107(1):123–150. Borjas, George J. 1994. "The Economics of Immigration." Journal of Economic Literature 32:1667–1717. Borjas, George J., and Stephen G. Bronars. 1991. "Immigration and the Family." Journal of Labor Economics 9(2):123–148. Borjas, George J., and Richard B. Freeman, eds. 1992. Immigration and the Work Force: Economic Consequences for the United States and Source Areas. Chicago: University of Chicago Press. Borjas, George J., and Lynette Hilton. 1996. "Immigration and the Welfare State: Immigrant Participation in Means-Tested Entitlement Programs." Quarterly Journal of Economics 111:575–604. Borjas, George J., and Stephen J. Trejo. 1991. "Immigrant Participation in the Welfare System." Industrial and Labor Relations Review 44(2):195–211. Boskin, Michael J., Lawrence J. Kotlikoff, Douglas J. Puffert, and John B. Shoven. 1987. "Social Security: A Financial Appraisal Across and Within Generations." National Tax Journal 40(3):19–34. Card, David, and Alan Krueger. 1992. "Does School Quality Matter? Returns to Education and the Characteristics of Public Schools in the United States." Journal of Political Economy 100:1–40. Chavez, Linda. 1992. "Commentary on Part Three." Pp. 297–299 in Immigration, Language, and Ethnicity, Barry R. Chiswick, ed. Washington, D.C.: American Enterprise Institute. Chiswick, Barry R. 1978. "The Effect of Americanization on the Earnings of Foreign-Born Men." Journal of Political Economy 86(5):897–921. Chiswick, Barry R., ed. 1992. Immigration, Language, and Ethnicity: Canada and the United States. Washington D.C.: American Enterprise Institute. Clark, Rebecca L., Jeffrey S. Passel, Wendy N. Zimmerman, and Michael E. Fix, with Taynia L. Mann, and Rosalind E. Berkowitz. 1994. "Fiscal Impacts of Undocumented Aliens: Selected Estimates for Seven States." Unpublished document, The Urban Institute, Washington, D.C. Coleman, James S. 1966. Equality of Educational Opportunity. Washington D.C.: U.S. Government Printing Office. Evans, W., Wallace Oates, and Robert Schwab. 1992. "Measuring Peer Group Effects: A Study of Teenage Behavior." Journal of Political Economy 100:968–991. Funkhouser, Edward, and Stephen J. Trejo. 1995. "The Labor Market Skills of Recent Male Immigrants: Evidence from the Current Population Surveys." Industrial and Labor Relations Review 48(4):792–811. Garrett, Daniel M. 1995. "The Effects of Differential Mortality Rates on the Progressivity of Social Security." Economic Inquiry 33:457–475. Garvey, Deborah L., and Thomas J. Espenshade. 1996. "State and Local Fiscal Impacts of New Jersey's Immigrant and Native Households." Unpublished document, Princeton University, Princeton, N.J.

OCR for page 13
The Immigration Debate: Studies on the Economic, Demographic, and Fiscal Effects of Immigration Hamilton, B. 1975. "Zoning and Property Taxes in a System of Local Governments." Urban Studies. 12:205–211. Hanushek, Eric. 1986. "The Economics of Schooling: Production and Efficiency in Public Schools." Journal of Economic Literature 24:1147–1117. Heckman, James, A. Layne-Farrar, and P. Todd. 1996. "Does Measured School Quality Really Matter? Understanding the Empirical and Economic Foundations of the Evidence." Working paper, University of Chicago, Chicago, Ill. Henderson, Vernon, Peter Mieszkowski, and Y. Sauvageau. 1978. "Peer Group Effects and Educational Production Functions." Journal of Public Economics 10:97–106. Huddle, Donald. 1993. "The Costs of Immigration." Unpublished document, Rice University, Houston, Texas. Huddle, Donald. 1994. "The Net Costs of Immigration to Texas." Unpublished document, Rice University, Houston, Texas. Kermit, D. 1992. "Does Marriage Make Men More Productive?" Economic Research Center/NORC Population Research Center Discussion Paper 38:92-92. Manski, Charles. 1993. "Identification of Endogenous Social Effects: The Reflection Problem." Review of Economic Studies 60:531–542. McManus, Walter S. 1985. "Labor Market Costs of Language Disparity: An Interpretation of Hispanic Earnings Differences." American Economic Review 75(4):818–827. Mieszkowski, Peter, and George Zodrow. 1983. "The Incidence of the Property Tax: The Benefit View versus the New View." In Local Provision of Public Services: The Tiebout Model after 25 Years, G. Zodrow, ed. New York: Academic Press. Neumark, David B., and Sanders D. Korenman. 1988. "Does Marriage Really Make Men More Productive?" Working paper, Board of Governors of the Federal Reserve System, Finance and Economics Discussion Paper Series 29, p. 59. Parker, Richard A., and Louis M. Rea. 1993. "Illegal Immigration in San Diego County: An Analysis of Costs and Revenues." Unpublished document, California State Senate Special Committee on Border Issues, Sacramento. Passel, Jeffrey S. 1994. "Immigrants and Taxes: A Reappraisal of Huddle's 'The Cost of Immigrants.'" Unpublished document, The Urban Institute, Washington, D.C. Pechman, Joseph A. 1985. Who Paid the Taxes, 1966–85. Washington, D.C.: The Brookings Institution. Pozo, Susan, ed. 1986. 1986 Essays on Legal and Illegal Immigration. Papers presented in a seminar series conducted by the Department of Economics at Western Michigan University. Kalamazoo, Mich: W.E. Upjohn Institute for Employment Research. Ramos, Fernando A. 1992. "Out-Migration and Return Migration of Puerto Ricans." Pp. 49–66 in Immigration and the Work Force, George Borjas and Richard Freeman, eds. Chicago: University of Chicago Press.

OCR for page 13
The Immigration Debate: Studies on the Economic, Demographic, and Fiscal Effects of Immigration Romero, Phillip J., Andrew J. Chang, and Theresa Parker. 1994. "Shifting the Costs of a Failed Federal Policy: The Net Fiscal Impact of Illegal Immigrants in California." Unpublished document, California Governor's Office of Planning and Research, Sacramento. Rothman, Eric S., and Thomas J. Espenshade. 1992. "Fiscal Impacts of Immigration to the United States." Population Index 58(3):381–415. Simon, Julian L. 1984. "Immigrants, Taxes, and Welfare in the United States." Population and Development Review 10(1):55–69. Simon, Julian L. 1989. The Economic Consequences of Immigration. Cambridge, Mass: Basil Blackwell. Sowell, Thomas. 1981. Ethnic America: A History. New York: Basic Books. Sowell, Thomas. 1994. Race and Culture: A World View. New York: Harper Collins, Basic Books. Stewart, William, F. Gascoigne, Mark Bannister, R. Wayne et. al. 1992. "Impact of Undocumented Persons and Other Immigrants on Costs, Revenues, and Services in Los Angeles County." Unpublished document, Los Angeles County ISD, Los Angeles, Calif. Summers, A., and B. Wolfe. 1977. "Do Schools Make a Difference?" American Economic Review 67(4):639–652. U.S. Immigration and Naturalization Service. 1987. Statistical Yearbook of the Immigration and Naturalization Service, 1986. Washington, D.C.: U.S. Government Printing Office. U.S. Immigration and Naturalization Service. 1996. Statistical Yearbook of the Immigration and Naturalization Service, 1994. Washington, D.C.: U.S. Government Printing Office. Warren, Robert, and Jennifer M. Peck. 1980. "Foreign-Born Emigration from the United States: 1960–1970." Demography 17(1):71–84.