July 2005

By Kelly O'Donnell and Kate Krause
Department of Economics, University of New Mexico

Acknowledgements

This project was funded by a grant to the New Mexico Advocates for Children and Families by the W.K. Kellogg Community Voices – New Mexico’s Shared Solutions, Grant Number P-0060131.

Laird Graiser and Bo Olcott of the New Mexico Taxation and Revenue Department provided essential assistance in obtaining and interpreting the tax data. This project could not have been completed without their assistance or the support of Kay Monaco and Robert Greenberg of the New Mexico Advocates for Children and Families.

Abstract

Absolute and relative poverty are pressing problems in New Mexico. Based on average poverty rates over the years 1995, 1996 and 1997, the U.S. Census Bureau reports that the state of New Mexico had the highest rate of poverty of any state in the nation. Children in particular suffer the long-term effects of growing up in poverty. This paper presents the results of simulations conducted for several proposed tax-based programs designed to decrease child poverty in New Mexico. Using a model constructed from individual tax return data, existing tax regulations and safety net programs for the poor, policies are analyzed and compared on the basis of cost (tax expenditure) and expected effects on poor families’ economic status. The results of these simulations lead to a prioritized set of policies that are most likely to cost-effectively reduce child poverty in New Mexico. In comparisons among these policies, a tax change that combines a revision of the existing state Low Income Comprehensive Tax Rebate with a tax credit specifically directed at children six years old and younger emerged as the most promising policy.

State Tax Policy and Child Poverty in New Mexico

Poverty is directly associated with negative consequences at all stages of life, and markedly compromises children’s health, educational opportunities and adult outcomes. Young children are particularly vulnerable. Poverty is directly linked to poor childhood health and limited educational opportunities, which in turn can adversely affect a child for the rest of his or her life. Impoverished children:

• are at increased risk for serious and chronic health problems,
• face higher infant and childhood mortality rates and greater exposure to violence in their homes and neighborhoods,
• begin school with more cognitive and development delays that compromise school readiness and
• drop out of school at higher rates than do children from wealthier families.1

The adverse effects of poverty are associated both with absolute wealth levels and with relative wealth, or the disparity between rich and poor. Interstate comparisons (for example, Kawachi, Kennedy, Lochner, and Prothrow-Stith, 1997) have demonstrated that this disparity between rich and poor negatively affects health status, including child health and infant mortality. Striking differences in the rates of absolute and relative poverty exist among states. Poverty generally, and child poverty specifically, are unevenly distributed within our country. Unfortunately, New Mexico has both a high rate of absolute poverty and a high degree of disparity between rich and poor.2 Thirty-four percent of New Mexican children under the age of six live in poverty.3

Although poverty reflects systemic issues that are part of a national (and international) structure, part of the differences in child poverty rates among states must be due to differences in policies and programs at the state level. If those differences were understood in terms of specific programs that reduced relative and absolute poverty and reduced child poverty, individual states, including New Mexico, could knowledgeably implement programs that addressed this serious problem. In this paper we demonstrate how a particularly rich data set, all tax returns filed in the state, can be used to model anti-poverty policies. While our data and the details of the state tax institutions used in the model are specific to the state of New Mexico, the methodology described in this paper is generalizable to poverty alleviation in other states.

The primary goal of this project is to model the effects of proposed tax-based policies designed to decrease child poverty in New Mexico. We specifically focus on tax-based changes for several reasons. First, at a national level, the EITC has been shown to be particularly effective at raising families from poverty (Greenstein and Shapiro, 1998) and encouraging them to remain employed (Meyer and Rosenbaum, 1998). Second, New Mexico has a unique tax-based policy already in place. Most poor New Mexicans file a tax return in order to take advantage of the Low Income Comprehensive Tax Rebate (LICTR). As a result, the target beneficiaries are identified within the existing tax base. These families are familiar with the tax system, and would be able to readily access benefits that are tied to that existing system. Third, assistance provided through the tax system avoids some of the pitfalls associated with other poverty relief programs. Eligibility is ascertained, and benefits delivered, by mail making it easy for residents in remote parts of the state to obtain the benefits for which they are eligible. This feature also eliminates the possible stigma that recipients report when required to prove eligibility in meetings with social workers and other benefit providers (see Moffitt, 1983 for a discussion of the deterrent effects of stigma). Finally, as a practical matter, use of an existing tax structure makes implementation administratively simpler and increases the likelihood of political support.

We use data provided by the New Mexico Taxation and Revenue Department to simulate the effects of several proposed tax-based initiatives. The proposed policies are specifically designed to lift families with children from poverty and to reduce the poverty gap, or the difference between a family’s income and the poverty threshold. Policies are analyzed and compared on the basis of cost (tax expenditure) and expected effects on poor families’ economic status.

The simulation model calculates poverty rates based on broadly defined income. Our income measure includes benefits under other safety net programs, including Temporary Assistance to Needy Families (TANF) and the Federal Earned Income Tax Credit (EITC). Food Stamp eligibility is not included as a benefit because it is an in-kind transfer and is substantially under-utilized. Comparing our calculations of eligibility based on tax return data with actual food stamp use reported in the New Mexico Human Services Monthly Statistical Report (October 1998), we find that only 28.6% of households that were eligible for Food Stamps actually received them. This low utilization suggests that some of the benefits attributed to a tax-based policy, such as convenience and reduction of stigma, are significant.

Inclusion of TANF and EITC income allows us to consider the relative effectiveness of proposed tax policies when those policies interact with eligibility for other income sources. The results of this analysis lead us to a prioritized set of policies and programs that are most likely to cost-effectively reduce child poverty in New Mexico.

Unique Characteristics of New Mexican Poverty and New Mexico’s Tax System

Poverty is not only more prevalent in New Mexico than it is in most other states, it is also different in at least three important ways. First, many poor New Mexicans live in remote rural locations, far from jobs and social service offices. Second, the income distribution in New Mexico is more unequal than in most other states. While New Mexico is home to many very poor families, it is also home to some very wealthy ones. Third, computation of numbers of families living below the poverty threshold masks the severity of New Mexico’s poverty problem. Almost half of New Mexican families who live below the threshold have incomes that are less than half that threshold. Nationally, 41% of people living in poverty are "severely poor’, defined as having an income of less than half the applicable poverty threshold.4 In New Mexico, an estimated 45% of poor people are severely poor.5 Policies that are capable of lifting families close to the margin over that threshold do not lift the poorest of the poor out of poverty. Fortunately, several aspects of New Mexican tax policy lend themselves to tax-based programs that address poverty.

New Mexico is characterized by a geographically dispersed population with minimal urbanization. New Mexico is the fifth largest state in the United States in area, but, as of July 1999, the state ranked 37th in population with fewer than two million people.6 Nationally, poverty is largely an urban phenomenon. In 1997, the Census Bureau estimated that only 23.3% of people living in poverty lived outside of metropolitan areas.7 In contrast, in New Mexico more than half of those living in poverty live outside a metropolitan area. Table 1 summarizes the incidence of poverty by family type and residence.

New Mexico is also characterized by high variance in poverty rates among counties. While New Mexico’s poverty rate is among the worst, if not the worst, in the nation (depending on calculation method) it ranks near the top in measures of wealth inequality. Larin and McNichol (1997) use Census data to compare the average income of families in the top quintile of the income distribution with the average income of families in the bottom quintile. For the years 1994-1996, New Mexico had the third highest difference among the states. Average income among the top quintile was more than fourteen times average income in the lowest quintile.9

According to 1995 Census Bureau estimates, 20.2% of New Mexicans, and 32% of New Mexicans under the age of eighteen, lived in households with incomes below the poverty threshold. That same year, only 3.0% of Los Alamos County residents under age 18 lived in poverty, while in Luna County, 49.1% of the residents under age 18 lived in poverty.10 (County-by-county poverty rates are provided in the Appendix.) Given this diversity, for each household type we calculate the distribution of the effects by whether the household resides in a small municipality, a large municipality, or a rural area. County-by-county effects are tabulated and presented in the Appendix.

The focus of this paper is on tax-based policies because the state personal income tax system is likely to be a particularly effective tool for the alleviation of poverty in New Mexico. The state of New Mexico currently administers three refundable tax credits through its personal income tax system. The child day care credit refunds a percentage of annual day care expenditures to low income working families, and the property tax rebate refunds a percentage of property tax or rent payments to low income filers who are over sixty-five. All households with modified gross income less than $22,00011 are eligible for the Low Income Comprehensive Tax Rebate (LICTR), including those with no earned income. In 1998, New Mexico returned roughly $25.3 million to low income state residents through LICTR.12

The existing state income tax system is unique in that virtually all poor New Mexicans, regardless of income, qualify for LICTR. The non-trivial sums of money distributed and the anonymous, non-stigmatizing nature of the delivery mechanism mean that New Mexicans who are geographically isolated or otherwise unable or unwilling to avail themselves of assistance from the Human Services Department do file state personal income tax returns. Estimates based on comparisons of the 1990 census with 1990 state tax filings indicate that 90 to 95% of poor New Mexicans file a tax return. Because LICTR has undergone two significant expansions since 1990, participation rates are likely to have increased. Coverage of the poverty population by other state-administered programs such as TANF and food stamps is not nearly as complete.

The tax returns filed by LICTR-eligible households are a rich source of data. In applying for LICTR, filers must complete the Personal Income Tax – Rebates and Credits (PIT RC) form. This form asks for detailed information on non-taxable income that is not required on the standard tax return. The Modified Gross Income (MGI) upon which LICTR refunds are based includes all components of Adjusted Gross Income (AGI) as well as Temporary Assistance for Needy Families (TANF), Supplemental Security Income (SSI), gifts, pensions and workers’ compensation payments. As a result, MGI constitutes a fairly comprehensive measure of both earned and unearned income.

Methodology

The comprehensive nature of LICTR eligibility and the information required by the PIT RC form allows us to use data derived from state income tax returns to draw inferences about tax policy effects on poor families. Data for this project was obtained from all 1998 state tax returns. This data set includes roughly 95% of the low-income population, allowing us to make accurate inferences about the revenue implications of specific changes, and to examine the distributional effects of policy on different sub-groups of the population.

The data were used to construct the 1998 New Mexico Personal Income Tax and Poverty Model. This model incorporates information provided on the PIT RC to estimate poverty status and eligibility for all state-administered entitlement programs as well as all state and federal tax rebate and credit programs. All calculations are based on statutory and administrative guidelines for determining program eligibility and anecdotal data provided by field workers on how written guidelines are actually interpreted. The detail on sources of income provided on the PIT RC allows us to explicitly account for earned income, child care, and housing cost disregards in establishing eligibility for Human Services Department administered programs such as Food Stamps, Medicaid and TANF.

This model allows us to simulate specific tax policy changes, thereby determining aggregate effects on tax revenue and disaggregated effects on households of different types. The model differentiates among several different household types: single parent, two parent, and families that differ by number of children. These simulations allow us to determine the size of the population eligible for a particular tax preference or set of preferences and the distribution of benefits across different household types within that population.

Policies Directed at Heads-of-Household

We begin our analysis by considering the effects of a change in the treatment of heads of households. One goal of the income tax rules is to adjust tax liability to reflect relative ability to pay. Rules governing head-of-household filing status address that goal. For example, taxes are assumed to be more burdensome to people with dependents than to those without dependents. Therefore, to enhance vertical equity, head-of-household filing status confers additional tax benefits on single filers with dependents. Head-of-household standard deductions and rate schedules are typically more favorable than those pertaining to single filers and somewhat less favorable than those pertaining to married filers.13 Tax treatment of heads-of-household has important implications for child poverty because the majority of poor children live in single-parent homes and are disproportionately represented in female-headed households with no husband present.14

New Mexico currently allows head-of-household filers a standard deduction of $6,250. This is more favorable than the $4,250 standard deduction allowed single filers and less favorable than the $7,100 standard deduction allowed married filers. The economic and distributional impacts of allowing head-of-household filers the married, and the married, filing joint, standard deduction and tax brackets are summarized in Table 2. Standard deduction changes would not affect child poverty rates because New Mexico does not tax households with income below the poverty threshold. This analysis assumes no compensating change in rates or other deduction amounts.

Households with income less than 140% of the poverty threshold have no tax liability. This change would not affect those families, and so would not lift any families from poverty. The average pre-LICTR state tax liability of head-of-household filers in our sample (those below 2.8 times the federal poverty threshold) is $80. The increased standard deduction decreases this population’s average pre-LICTR tax liability to $70, a change of ten dollars. Head-of-household tax brackets are also slightly less favorable than those pertaining to married couples filing joint returns. Making head-of-household standard deductions and tax brackets the same as those for married, filing joint would decrease average pre-LICTR tax liability by fourteen dollars.

While the tax expenditure for these proposals would be fairly low ($2.2 million and $2.9 million respectively) these changes do not target poor families exclusively and have no effect on families who have incomes below the poverty threshold. Less than 50% of the total tax expenditure would accrue to families with incomes less than 280% of the poverty threshold. We therefore focus on policies that most directly effect poor children.

Policies directly targeting children

Nationally, the poverty rate among the elderly has declined dramatically and the poverty rate among working age adults has remained fairly constant, while the poverty rate among children, and particularly among young children, has increased. In 1966, the first year for which data disaggregated by age is available, 17.6% of children in the United States lived in poverty, 10.5% of adults aged 18 to 64 lived in poverty, and 28.5% of those 65 and older lived in poverty. By 1997, 19.9% of children (21.6% of those under age six), 10.9% of working age adults, and 10.5% of those 65 and older lived in poverty.15 New Mexico’s situation is more disheartening than that of the country as a whole. New Mexico’s poverty rate, calculated as a three-year average covering the years 1995-97 is, at 24.0%, last among the fifty states and the District of Columbia.16 Approximately thirty-two percent of New Mexican children live in poverty17 and approximately half of those live in extreme poverty, defined as having income less than one-half the poverty threshold.18 Poverty rates are even higher among the youngest children. Although the Census Bureau does not calculate poverty rates for this age group, the National Center for Children in Poverty estimates that 36.8% of New Mexican children under the age of three live in poverty.19

Given the severity of this situation, and the long-term consequences of growing up in a poor household, we consider several policies that are intended to alleviate poverty in families with children. The economic and distributional effects of these policies are summarized on Table 3. Policies that we consider are:

• Providing for an additional tax credit for each child living in a family with income less than the poverty threshold;
• Instituting a graduated refundable child credit;
• Replacing LICTR with a graduated child credit;
• Establishing a Family Income Tax Credit similar to one available in Arizona;
  Revising LICTR to provide for 2 extra exemptions per child and increasing the rebate for filers with more than one exemption and having more than $3500 in Modified Gross Income (MGI); and
• The LICTR revisions described in number 5 together with providing for an additional credit for families with children under age 6, which we call the Early Childhood Advantage Tax Rebate.

Non-Graduated Tax Credits

Perhaps the most straightforward way to address child poverty is to give money to poor families that have children. Therefore, the first policy to be considered is the establishment of a tax credit for each child in families with MGI less than the poverty threshold.20 We consider two levels of tax credits, one in which the tax credit is equal to $300 per child and one in which the credit is equal to $400 per child.

A $300 credit would cost the state of New Mexico approximately 45.6 million dollars, while a $400 credit would cost the state approximately 60.8 million dollars. The $400 credit would lift just over five thousand families, including 11,827 children, from poverty. This compares with 3,767 families, including 8,832 children, lifted by a $300 credit. The average cost per child lifted is approximately $5,000 for each policy. The two plans differ most significantly in their success at reducing the average depth of poverty. To determine depth of poverty, we calculate the difference between the applicable poverty threshold and each family’s actual MGI, and then average that difference over all families living in poverty. The reduction in depth of poverty is the difference between this average without the policy and this average with the policy in place.21 A $400 credit reduces average depth of poverty by 27% while the $300 credit would reduce depth by approximately half that amount. In addition, the $400 credit reduces depth more efficiently. Each percentage point decrease in depth costs, on average, approximately $2.2 million with the $400 credit. One percentage point decrease in depth with the $300 credit costs, on average, approximately $3.5 million.

This level of aggregation masks differences across families. The effects of the policies on families of different sizes are detailed on Table 4. As would be expected, the families that would benefit most from these policies are those with the most children. For example, while families with four or more children constitute only 7% of all poor families, they constitute 16% of the families that would be lifted from poverty by a $400 per child credit.

A policy with this distributional effect may be unpalatable to legislators who are concerned about fertility incentive effects of the credit. That is, some may think that this policy would tend to encourage large families. As can be seen in the last column of Table 4, this effect is reduced with the graduated credit to which we now turn.

A Graduated Refundable Tax Credit

The constant credit described above provides the same benefit to families regardless of their depth of poverty. Those families lifted from poverty are those who are closest to the threshold prior to institution of the credit. To address the particularly grim situation faced by families in extreme poverty, we consider a graduated, refundable tax credit. The credit would be greatest for families in extreme poverty, and would decrease as Modified Gross Income (MGI) approached the poverty threshold. The policy we simulate provides for a refundable tax credit equal to the per child amount times the number of children up to a maximum of four children. Credit amounts are as shown on Table 5.

If this policy were adopted, the tax expenditure would be nearly $60.5 million dollars, or just less than the cost of the $400 per child credit. Just over seventy-five thousand families would benefit and 3,642 families would be lifted from poverty, significantly less than the 5,002 families that would be lifted by a constant $400 credit. More than eight thousand children would be lifted from poverty at an average cost of $7,404 per child. The graduated credit would reduce the average depth of poverty by 17%. Although the graduated credit provides larger credits to the poorest families, its effect on depth of poverty compares unfavorably with the $400 credit, which reduces the average depth of poverty by 27%. The graduated credit would lift approximately the same number of families and children from poverty as would a $300 credit, but at a substantially higher cost. It would yield a slightly greater reduction in the depth of poverty (17% vs. 13%) at a similar cost per percentage point decrease.

The structure of this policy leads to aggregate results that appear to suggest that a graduated credit would deliver fewer benefits per dollar of tax expenditure than would a flat rate. However, the credit specifically targets families in extreme poverty. Benefits are concentrated among those whose MGI is so far below the threshold that even a substantial credit is inadequate to lift them from poverty. Similarly, the reduction in overall average depth of poverty is not as impressive as with the flat credit. However, that reduction is distributed so that the poorest families would experience the largest reductions in depth of poverty while those families whose incomes are nearer the threshold would experience smaller reductions in depth. It is likely that families living in extreme poverty experience the most damaging effects of poverty. Therefore, a policy that delivers a smaller aggregate benefit, but delivers greater benefits to those who most need them, may be a more cost-effective policy.

The last column of Table 4 illustrates the effect of this policy change on families by size of family. While one-child families constitute more than forty percent of the poor population, fewer than one-fourth of the families lifted from poverty by this policy are one-child families. A higher percentage of families with two or three children would be lifted from poverty, presumably because each child yields a credit. The distribution of families lifted from poverty relative to the overall distribution of poor families did not vary by whether the family lived in a large city, a small city or a rural area.22

Replacement of LICTR with Graduated Tax Credits

One way to reduce the tax expenditure associated with tax credits would be to eliminate LICTR for families with MGI less than 100% of poverty and replace it with the graduated child credit described above. Eliminating LICTR for recipients of the child credit (all families beneath 100% of poverty) would decrease the tax cost of the graduated child credit by more than ten million dollars. LICTR, as it pertains to households without children and those with incomes above the poverty threshold, would remain unchanged.

This policy would benefit the same 75,010 families as would the tax credits, but would lift fewer of those families from poverty: 2,992 compared with more than 3,500 for the graduated and $300 credit and more than 5,000 with the $400 credit. The annual cost per child lifted from poverty would be $7,251, similar to the per child cost of the graduated credit, but more than the approximately $5,000 cost associated with either of the proposed constant-rate credits. The reduction in average depth of poverty would be 14%, similar to that obtained with a $300 credit, and for a similar cost per percentage point reduction (approximately $3.5 million). This compares unfavorably with the $400 credit, which reduces the depth of poverty by 27% at a cost of $2.24 million per percentage point reduction.

Family Income Tax Credit

The Arizona Family Income Tax Credit is a non-refundable credit equal to the lesser of 40 times the size of the household or $240 for married filing joint and head of household or $140 for married filing separate or single filers. The size of the household is the sum of personal exemptions and dependants. Eligibility is based on income and filing status as shown in Table 6.

The effect of the Arizona credit is to eliminate tax liability for filers below the poverty thresholds. The amount of the credit does not vary with income. However, the income thresholds are a function of household size. The Arizona Family Income Tax Credit does not give preference to the elderly or the blind as does LICTR. The policy discussed in this section is a refundable personal income tax credit, but otherwise identical to the Arizona Family Income Tax Credit. As with the other simulations, baseline poverty status for purposes of policy analysis and comparison is based on Modified Gross Income.

At $23.2 million, the cost of this policy is less than half the cost of the graduated credit (with or without LICTR) or the $400 credit, and is more than twenty million dollars less than the $300 credit. Just over twice the number of families, 153,305 compared with 75,010, would benefit. However, only 852 families and 1,803 children would be lifted from poverty by this policy. The annual cost per child lifted, $12,868, is more than twice the cost per child associated with a $300 or $400 credit, and also compares unfavorably with the cost per child associated with the graduated credit ($7,404). The average depth of poverty would be reduced by only 4.6% at a cost per percentage point decrease of approximately five million dollars. The distribution of families that would be lifted from poverty is similar to the overall distribution of poor families with respect to family size and residence. Replacing LICTR with a New Mexico personal income tax credit that is refundable but otherwise identical to the Arizona Family Income Tax Credit would save the state $2.1 million and increase child poverty by 21 households.

LICTR Revisions Combined with Early Childhood Advantage

The last policy that we consider is a two-tiered proposal that combines restructuring LICTR with a credit targeted specifically at children aged six and under. There are many advantages to using the existing LICTR as a vehicle for implementation. Familiarity among legislators and among recipients increases the chance that the policies will be enacted and that they will be used. The administrative costs and start-up investments are minimized because collection and enforcement is already in place. For example, the Early Childhood Advantage Tax Credit could be implemented with literally no changes in the state tax rate tables.

We specifically target the under-six age group for two reasons. First, this age group is particularly vulnerable to the negative health effects of poverty, and alleviating those effects would yield long-term benefits. Second, the poverty rate is particularly high among families with pre-school children. Under this policy, LICTR would be revised to allow poor families with children two extra exemptions for each dependent child. This would make the exemption status of children equivalent to that of the elderly and the blind. In addition, the LICTR rebate would be increased by 20% for filers having more than one exemption and more than $3500 in MGI. These specific provisions are designed to encourage work in two ways. First, the increased rebate is only available to filers with some earned income. Second, by increasing the rebate for the working poor, the policy would ameliorate the effects of the phasing out of other entitlements associated with earning income. For example, working poor families lose child care subsidies and Medicaid eligibility as their earned income increases. The rebate would offset these lost benefits. These changes would not reduce LICTR rebates to any current LICTR beneficiaries.

The Early Childhood Advantage Tax Rebate would provide an additional refundable personal income tax credit to families with children six years old and younger. This credit would be administered in conjunction with LICTR, either as revised as described in the preceding paragraph or as it currently exists. The amount of the Early Childhood Advantage credits would vary with MGI and the number of young children present. Specific amounts are given in Table 7.

The combined tax expenditure for these programs is twenty-nine million dollars, approximately half the cost of either the graduated credit or the $400 credit. More than one hundred thousand families would benefit, three thousand of which would be lifted from poverty. More than six thousand children would be lifted from poverty by these policies at an average cost of approximately $2,500.23 This is less than half the average cost associated with either of the constant-value credits and approximately one-third the average cost associated with the graduated credit (with or without LICTR). The average depth of poverty would be reduced by 9% at a cost per percentage point reduction of approximately $1.7 million, far less than any other policy considered.

The LICTR revisions alone would not be as cost-effective as the combined policy proposed. The cost of the LICTR component is $13.6 million. This policy alone would lift only 446 families from poverty, including 734 children, at an average cost per child lifted of $14,387, the highest average among all policies considered. In addition, the depth of poverty would be decreased by only 4.5%, slightly less than the decrease that would be obtained with a Family Income Tax Credit. Because it is so narrowly focussed, adding the Early Childhood Advantage component adds $15.3 million in tax expenditures, but increases the number of families lifted from poverty by more than six-fold and the number of children lifted by more than eight-fold.

Lower Cost Alternatives

Political realities force us to consider alternative policies that can be implemented with a lower tax expenditure. Individual components of the combined policy could be instituted alone, although the effectiveness of the policies would be reduced. Credit amounts could be reduced, and this cost-saving modification is described at the end of this section.

A LICTR revision that would give children the same number of exemptions as the elderly and blind (two extra exemptions) would cost approximately $7.2 million and has several attractive features. The policy specifically targets poor children, can be administered without changing the structure of LICTR or the existing tax tables, and brings children to the same status as other groups that are particularly vulnerable to the negative consequences of poverty. However, this revision alone does not incorporate the work incentive associated with increased rebate rates at MGI greater than $3500.

In addition, the LICTR revision does not address the important issue of poverty among the youngest children in the state. Providing for a third exemption for children six and under would add approximately one million dollars to the cost. This additional cost is relatively low, as would be its effect on poor families. The current LICTR schedule reaches a maximum at six exemptions, and so an additional exemption for younger children would not yield additional benefits for any family that already has six or more exemptions. Amendment of the tables to allow more exemptions is possible but would require longer lead-time and higher administrative costs.

Approximately sixty thousand New Mexican children aged six and under live in poverty. An outright grant of just sixteen dollars to each of those children would cost almost one million dollars. A total tax expenditure of ten million dollars could purchase the two additional exemptions described above plus a credit of less than fifty dollars ($46.67) for each poor child under the age of seven. These are token amounts and are not likely to significantly change the lives of these young children. Therefore, the value of either an additional exemption or a tax credit is largely symbolic. However, enactment of this policy would constitute political acknowledgement of the needs of young children and would seed the debate that may lead to significant improvements in their lives.

Policy Recommendations

Among the policies considered, the combined LICTR revision and Early Childhood Advantage Tax Rebate offers the most cost-effective tax-based program for the alleviation of poverty among New Mexican children. Furthermore, the tax expenditure associated with these policies is likely to be over-stated. New Mexico has an extremely broad-based gross receipts tax. Virtually every dollar returned to poor New Mexicans will be spent on goods or services that yield gross receipts tax revenue. As a result, roughly $1.3 million of the program’s estimated $29 million annual cost would be returned to the state each year.

This "best" policy is divisible into individual components. For just over seven million dollars, the state of New Mexico could grant children the same exemption status it grants the elderly and the blind. This policy change is equitable, efficiently uses the existing tax structure and tables, and provides a direct benefit to all poor families with children. It does not address the serious concerns associated with poverty among the very young. Two additional policies were considered that directly address these concerns: an additional exemption and a credit for each young child. These policies would have a minimal impact within politically and economically feasible tax expenditure ranges. However, either one would be an important step in the direction of acknowledging the importance of alleviating poverty in this vulnerable population.

Conclusion

Absolute and relative poverty are pressing problems in New Mexico, and children in particular suffer the long-term effects of growing up in poverty. By some calculations, New Mexico has the highest rate of poverty of any state in the nation. New Mexico’s situation is exacerbated by depth of poverty and income inequality among its residents. Poverty rates calculated as the proportion of families with income below the poverty threshold mask problems associated with depth of poverty and income disparities. The extremely poor face worse consequences than those living just below the threshold and are less likely to be lifted from poverty by existing programs like the federal Earned Income Tax Credit. Wealth in New Mexico is distributed more unequally than it is in most other states. Even when controlling for differences in absolute poverty, this relative poverty has been shown to be associated with worse health outcomes.

While poverty and inequality in New Mexico are particularly severe, there are also unique opportunities within the state tax system that address these problems. The state has already implemented tax-based policies that reach nearly every poor family in the state. The most comprehensive of those policies, LICTR, provides benefits to families without the stigma and travel costs associated with other social services. This is especially important in a state in which poor families are geographically dispersed.

Using detailed tax return data from more than 95% of the poor population, we conducted simulations of several tax-based policies targeting poor families, and particularly those with young children. Our immediate goal was to identify policies that would alleviate the immediate and long-term problems associated with poverty in an affordable and politically palatable way. A secondary objective was to demonstrate the usefulness of using existing state tax policy to alleviate poverty. The policy proposals presented in this paper were included because they could be administered through existing programs and could be readily accessed by families with children.

In comparisons among these policies, a tax change that combines a LICTR revision with a tax credit specifically directed at children six years old and younger emerged as the most promising policy. Policies that specifically target families with young children address the poorest and most vulnerable segment of the population. The benefits associated with reducing depth of poverty, and in some cases lifting these children out of poverty, extend beyond the direct numbers calculated by the simulation model. By investing in policies that reduce childhood poverty, the state would be investing in the future.