COMPARISONS OF REGIONAL REVENUES AND EXPENDITURES
The levels and distribution of wealth within the Greater Washington metropolitan area provide valuable insights into the region's prosperity and potential. In particular:
o financially, regional trends in income tax returns are informative,
o as are the regional trends in state and local revenues and expenditures from the middle '90s .
o more recent comparisons of FY02 budgets from each of the 'inner metro area' jurisdictions continue to show big variations in spending patterns between DC and its suburbs.
o But there is a special burden associated with the very high costs of poverty which impacts heavily on DC and is getting worse.
o An interesting poll in early 2001 of people both below and above the poverty line shows rather remarkable agreement on some of the causes and cures for poverty. There is a broader spread in views between Republicans and Democrats, and between white and black, than there is between the poor and those better off.
o 20-year trends in families in poverty tend to confirm these views, with black single moms with almost 3 kids each providing the largest single group.
o Census 2000 provides a useful update of 20-year trends in US poverty, indicating significant changes in who is poor and where they live.
o And a recent set of analyses from the Urban Institute documents the extent to which the cards are stacked against poorer kids as a result of "sociodemographics"; family stress, and family turbulence which lead to poor school performance and higher levels of difficulty outside the schools;
o There is an interesting new NARPAC analysis treating the newly released DCPS Facilities Modernization Plan indicating that an encouraging reduction in births to single moms will eventually result in a much smaller overall public school enrollment. If DC's public education problems get smaller numerically, perhaps they will be more readily fixed; o Until then there will be inescapably higher per capita costs of government because of DC's income mix, which includes an unusually large number of residents below the poverty line, many of whom still are children;
o as well as inescapably lower per capita revenues due to the persistent imbalance between black and white earning power: the difference between identical cities, one all-black, and the other all-white, is startling;
o and underlying all these problems is the staggering number of births by unwed mothers, which now exceeds 60% for Blacks, with Whites seemingly trying to close the gap!
o All the above suggest the potential need for some state payments in lieu of taxes to level the regional welfare burden by "poverty-sharing".
o Finally, comparisons of taxes on small business can provide some indication of each jurisdiction's ability to attract and keep new businesses.
o NARPAC has collected many of the factors addressed here and elsewhere in its separate briefing summary entitled Economic Challenges for DC;
Data on federal income taxes as tabulated in the IRS "Statistics of Income" reports, can be used to determine trends in "economic demography" in the Greater Washington Metropolitan Area. Comparative data are presented for the US as a whole, the District of Columbia, and its two immediate neighboring states, Maryland and Virginia. Because it is not possible to separate out what share of each state's wealth is part of this metro area, the reader will have to interpolate. Almost certainly, something more than 33% of each state's wealth is applicable. The complicated tables on the following pages present the data from 1988 and 1994, and the narrative below is intended to explain their general content.
For the 1994 tax year, some 116 million federal tax returns were filed with the IRS. DC taxpayers filed 281,000, while Maryland taxpayers submitted some 2,332,000 and Virginians, 2,981,000. Across the US, returns have increased some 5.6% in the seven years from 1988 through 1994. In Maryland, the increase has been only 2.3%, while in Virginia it has been 7.4%, indicating a more rapidly growing taxpayer base. By comparison, the total DC returns have dropped some 13.2%.
For purposes of analysis, the tax returns can be divided into those whose Adjusted Gross Income (AGI) is more or less than $50,000 annually. As incomes increase in real terms (and due to inflation), the number of returns in the higher category should increase. Hence, for the US as a whole, lower income returns have dropped 2.3%, while the higher returns have increased more than 59%. Nevertheless, almost 81% of returns remain in the lower category. In DC, the lower category has declined 18.8% in seven years, while the higher category has increased 26.7%, suggesting that the DC is not keeping up with the country at large. By comparison, Maryland lower income returns have decreased 7.5%, while the upper half grew 48.8%. In Virginia, the lower incomes dropped 1.6% while the upper incomes increased 56.7%, indicating that it is adding richer residents faster than Maryland.
Most exemptions are for children and spouses, while a few are for the blind and other disadvantaged. Americans claimed a total of 254 million exemptions in 1994, up 5.9% from 1988. Maryland's total rose 6.4%, and Virginia's 9.0%, while the DC's fell 10.3%, indicating a clear drop in family taxpayers in the nation's capital. Perhaps more interesting, the average exemptions per return are now 2.18 nationally, 2.12 in Maryland and Virginia, but only 1.89 in DC, again indicating smaller than average families.
The best available measure of taxpayer wealth is probably Adjusted Gross Income -- Line 31 on your federal income tax form. In 1994, Americans reported $3.9 trillion in income (on which they paid $565 billion--l4.5%- in income taxes). This was up over 27% since 1988. Maryland's AGI was up over 24%, and Virginia's almost 30%. The DC's was up also, but only 7.6%, indicating that it was still growing, but slipping relative to its suburbs. The share of the AGI produced by upper-end returns (over $50,000) rose almost 57% nationally. In DC, they rose less than 29%, while in Maryland they rose over 50%, and in Virginia over 58%.
The actual average AGI for the lower share rose almost 21% to $35,900 nationally, while it rose 24% to $37,300 in DC, and about 21% in each Maryland (to $38,600) and Virginia (to $35,900). In fact, there is not all that much difference in average wealth across the region, though the DC is still tops. The average AGIs for the lower share are also very nearly the same: $19,300 nationally, $19,600 in Maryland, $19,300 in Virginia, and $18,900 in DC. On average, the less well off taxpayers are not that different throughout this region.
The greater spread is in the upper share. The national average AGI for those reporting over $50,000 is now $97,600, compared to $94,400 in Maryland and $93,100 in Virginia. But in DC, that average is $122,000. These differences are more stark if the dividing line is moved up to $100,000 in AGI. Above that mark, some 4.9% of Americans report an average of $216,400--well above Maryland's $167,300 and Virginia's $181,800. But in DC, the top 5.4% report an average AGI of $245,700. In short, the wealthiest few in DC are far better off than their counterparts locally and nationally. It is hard to believe that they need tax relief!
AGI reflects directly in taxes paid to the Federal Government. The $565 billion paid nationwide in 1994 was up 32% from 1988, up 45% in Maryland, and up 32% in Virginia. Despite claims of a crumbling tax base in DC, federal tax revenues from the DC were still up 18% from 1988--no bonanza, but no calamity either. It is interesting to note, however, that 72% of all income taxes are paid nationally by those with AGIs about $50,000. The equivalent fraction is 75% in Maryland, 73% in Virginia, and 75% in DC. Compared to a national average of 43%, Maryland comes in at 38% and Virginia at 41%, for the share paid by those with AGIs over $100,000. In the DC, however, almost 54% of all federal taxes are paid by the top 5.4%. The percentage of returns reporting over $100,000 is 3.9% nationally, 5.6% in Maryland, 4.9% in Virginia, and 5.4% in DC. This region clearly has more rich taxpayers (294,000) than most areas, and those few in the District (15,300) are richer than their counterparts elsewhere.
The DC and its two neighboring states paid a total of $23.0 billion in federal taxes in 1988, and $29.4 billion in 1994. DC's share rose from $1.5 to $1.7 billion. The DC, despite its wealthy few, is not keeping pace with its neighboring states. DC's share of the region's total AGI has dropped from an already low 5.9% in 1988 to an even lower 5.1% in 1994, while the actual taxes paid have declined from 6.4% to 5.9% of the regional total in those seven years. Even if only one-third of each state's total wealth is related to the GWMSA, the DC would be responsible for less than 20% of the regional total. It is no wonder that the DC is having difficulty providing services to its neighbors with such a small share of the area's tax base. Even if it is not declining absolutely, it is certainly a small and diminishing part of its metropolitan area.
Data on the following tables are derived from the IRS Statistics of Income Reports, and then manipulated arithmetically to provide additional proportional information.
Statistics of Income data for 1995 have finally been reviewed by NARPAC, Inc. with the following results. 1995 was another good year economically for the US-- and for federal revenues. Federal tax returns increased by 2%, and exemptions only 1.6%. Adjusted Gross Income (AGI) rose a total of 7.3%, and federal revenues 10%, due to the progressive tax schedule: average AGI rose 5.2%, but the share of AGI from returns with AGI over $50,000 rose 12.1%.
In fact, the DC region (represented here by total tax returns for Maryland, Virginia, and the District) did not keep pace with the US average. Returns for the area rose 1.6%; AGI, 5.6%; and average AGI 4.3%. The area's average AGI of $38,700 is still well above the national norm of $35,200, but the advantage is declining. Claims that DC's tax base is eroding, are technically true, but the more serious problem is that it is not keeping up with its prospering suburbs.
Within the region, the District again fared the worst. Total returns dropped 3.5%, though those over $50K rose 3%. Total AGI dropped 0.2%, but AGI for those above $50K rose 1.5%. Average exemptions (mainly for kids) continued to decline slowly, and stay well below the national norm (1.87 v 2.17). The very wealthy in DC (average AGI above $100K) still represent a much larger share (6.0%) than for the nation as a whole (4.4%), but their average AGI of $236,700 is no longer so much higher than the national average of $220,800, but still very substantially above those for Maryland ($187,000) and Virginia ($185,300). NARPAC, Inc. has no particular insight into the reason for these differences, but suspects irreverently that DC has too many lawyers, while the region has too few industrial moguls.
NARPAC, Inc. assumes the close correlation between the number of DC taxpayers paying DC taxes and those paying federal taxes continues. In recent years, local tax revenues have been about 34% of those raised by the federal government, and less than 25% of those income taxes are now paid by households with income below $50,000.
This analysis offers a representative sampling of how states and local governments raise and spend their public monies, and compares that to equivalent data for the District of Columbia. The data are taken consistently from the Census Bureau report, Government Finances: 1990 91. Though the information may appear somewhat dated, the trends change very slowly from year to year.
For obvious reasons DC is compared to its neighboring states of Maryland and Virginia, and to Delaware as well (See endnote). Data from these three states are then combined into a single State of "Delmarva" simply to average out the idiosyncrasies of each. Several popular myths seem to be dispelled by the facts:
The popular assumption that states somehow underwrite the costs of their metro areas from someplace outside those metro areas is not supported. States do raise more of the total revenues than their local governments (counties, municipalities, school districts, etc.) by 63% to 37% (see Table 1 , but state taxes are paid primarily by metro area residents because over 85% of the personal wealth resides in the metro areas in "Delmarva" (82% nationwide). Over 80% of the people live in "Delmarva" metro areas (compared to 78% nationwide), and they are on average one-third richer than their country cousins--as they are nationwide. See Table 2 .
There are significant differences between the three states in how they raise their taxes (Delaware relies more on state taxes, Maryland depends on their counties), and how the funds are spent (Maryland relies most heavily on their counties; Virginia and Delaware rely more on "Other local", meaning their school districts). The Delmarva state authorities transfer 10-15% of their revenues to local authorities to spend.
Re-allocation of Federal Funds to States:
The most striking statistics, however, are shown on Table 3 comparing both per capita revenues and expenditures for DC with the Delmarva states collectively and separately. These numbers are totals for all levels of state governments. Per capita revenues required to run DC amounted to $8825 per resident in 1990-91, 2.2 times the $4000 per capita (pc) for Delmarva--varying from $4653 pc in Delaware to $3797 in Virginia. Per capita expenditures look even worse: $9376 pc for DC, compared to $3988 pc for Delmarva. There are no areas of revenue raising where DC lagged behind its neighbors. It should be noted that these statistics consistently exclude federal assistance directly to individuals (such as social security payments) or the administration of those programs.
There is also virtually no area where DC spent less than its neighbors-
except on higher education (40% less) and roads (60% less). The latter is
not surprising since the population density in Delmarva is 227 persons per
square mile, compared to 9951 in DC. In some areas DC spends twice as
much (for administration and debt interest); three times as much (for
social services); and four times as much (for public safety).
It would appear at first glance that the DC's problems have little to do with its inability to raise revenues, but rather with its inability to control spending. On closer inspection, however, it would appear that DC's root problem is one of an incomplete, isolated, inner city trying to function as a multi-level jurisdiction.
Table 4 illustrates the per capita state and local
DC, the Delmarva states separately and combined, and for the US as a
whole. The comparisons are instructive: DC per capita spending of $9376
is 2.35 times higher than the $3988 for the Delmarva states. However,
the Delmarva states are some 7% below the national norm of $4274 per
capita. If the Delmarva and DC expenditures are combined, then their
aggregate expenditures per capita of $4255 essentially matches the
national norm. Hence, by raising Delmarva per capita spending by $267,
the DC per capita spending could be reduced by $5121, a direct result of
spreading the inner city costs across a population base twenty times as
large. From the standpoint of leveling the playing field, the financial
advantages of total regionalization are very large indeed.
The prior sections indicate clearly that DC has been paying substantially more per capita than its surrounding jurisdictions, despite what can only be called a significantly lower quality of life at least for many of its residents. Much of that data dated back to the early 1990's. NARPAC has more recently updated these comparisons using data from the FY02 budget proposals of DC, and its five contingent neighbors: Alexandria City, and the counties of Arlington, and Fairfax in Virginia, and Montgomery and Prince George's in Maryland. The differences in total revenues and expenditures have moderated somewhat, in part as DC has given up some of its most expensive (and inappropriate) "state functions", and as increases in local government spending were limited by the existence of the DC Control Board, now scheduled to go out of business when the FY2002 budget year begins. NARPAC has also changed its evaluation yardstick from dollars per capita provided or spent by each resident, to dollars per household, which eliminates the biases of differing child populations. The newer comparisons are discussed below.
The major difference in DC's revenue and expenditure profiles compared to its suburbs, of course, is the lack of state-level organizations and spending for major items such as welfare on the one hand and state police on the other. NARPAC can only compute an average state cost per state household and add it to the local costs per local household. This may well introduce some inaccuracies but they should not detract from the overall conclusions (though they may make Montgomery County look somewhat out of line)
As illustrated on the upper chart below, the average DC household will consume some $21,600 in public resources, compared to an average of about $15,600 for the "inner suburbs". In fact, DC's locally-raised revenues are almost the same per household as the total expenditures of its suburbs from all sources; local, state, and federal. The additional DC costs are paid by the federal government from nationally-raised revenues. The fact remains, however, that District households require 25% more public funds than its surrounding neighborhoods.
The comparisons look somewhat different when measured by how many local and state government personnel are employed per household or per thousand households to get whole numbers. As indicated below, DC still 'wins' with the highest number of 133 government FTE's per thousand households (from all sources), but Alexandria is not far behind. However, the five suburbs together again average out to employing about three-quarters as many government personnel as DC. It should also be noted that the number of state government workers is at best one-third as large as the local force.
Lastly, NARPAC has singled out two areas and tried to compare the total spending at local and state levels for human services and public safety (less the state correction facilities). Here the differences are more stark, and explain in large measure the higher costs and personnel levels for DC. As shown below, DC ends up spending almost three times as much on human services as the inner suburban average ($8700 v $3100 per household), and two and a half times as much on 'public safety'(police, fire, and justice systems). As discussed elsewhere, DC has way more than its share of poor and disadvantaged and it also pays them substantially more per capita ($1750) than Maryland ($900) or Virginia ($500) according to DC's FY02 budget presentation.
THE VERY HIGH COSTS OF POVERTY--for DC Some interesting comparisons have recently come to light (January, 1999) indicating a very substantial drop in the number of persons living in poverty in 1996 compared to 1990 for each county in the metro area--a trend which has presumably continued. A Washington Post article, based on the work of DC's demographer George Grier (as reported by the Greater Washington Research Center), provides his estimates of the number of people (80,000) below the poverty line. NARPAC, Inc. has taken those data points and generated their own analysis of their meaning. However, Census estimates places this number much closer to 130,000--which would make these comparisons even worse.
Including DC and its five immediate neighboring counties/cities (Alexandria, Arlington, Fairfax, Montgomery, and Prince George's), the total number of people in poverty has dropped by some 63,000 persons (29%) since 1990. However, the statistics for DC are in stark contrast to those for its suburbs. While DC appears to have dropped by 17%--compared to 39% in the suburbs based on actual head count (not unimportant), the ability to pay is more accurately related to the change in poor compared to the total supporting population. By this measure, DC's per capita poor have dropped less than 5%, whereas the suburbs "failure rate", if you will, has dropped more than 43% as population increases.
But even more indicative of the burden of the poor on those paying taxes, is the ratio between welfare recipients and that share of the employed work force above the poverty line--assumed by NARPAC to be 85%. (The Tax Revision Committee indicates that 40,000 of DC's 272,000 tax returns are below the poverty line.) From this viewpoint the burden of each poor person is borne in DC by 2.6 non- poor workers. This ratio would be even worse if the Census estimates are correct. The average ratio for the five suburbs listed above is 17.8 non-poor workers per poor person--varying from 11.2 in Arlington to 20.5 in Montgomery County. Prince George's is near the average at 17.2. The result is that the burden of the poor on DC taxpayers is four to eight times higher than in the suburbs (6.8 times the weighted average).
[It should be noted that the Post article quotes 5.7 non-poor residents per poor resident, but this is meaningless since all the poor require assistance, while only the earning non-poor contribute to that assistance].
Earlier statistics have indicated that the average welfare payment in 1990 to DC recipients was about $3200, and could easily have risen to $3800 by 1996. But the higher costs of poverty on local governments is the cost of government workers employed to administer directly or indirectly to those in poverty. NARPAC estimates that perhaps 80% of Health and Human Services employees are dealing with the poor, while perhaps 25% of public schools, police and emergency services, and corrections personnel are also on the payroll to deal (indirectly) with those below the poverty level--and the consequences.
At DC's 1997 personnel levels, this amounts to some 8900 government employees, equating to eleven government personnel per 100 poor people. In fact, this number can also be used to rationalize why the total number of government workers per thousand residents is substantially higher in DC (65) compared to the total state and local employees per thousand residents in Maryland (50) or Virginia (54).
With an average DC city worker salary-and-benefits near $60,000 per year, almost $6700 per year is needed to administer to each poor person. Combined with welfare benefits, this approaches $10,000 per year to be paid by 2.6 workers (or less). Possibly more alarming, however, is the fact that the "indirect costs" of accommodating the poor (i.e., administration) consumes two-thirds of the cost, and only one-third reaches the poor as monetary subsidies.
Even if most of the welfare checks are paid by federal grants, a large share of the government personnel costs falls on DC taxpayers. Given that per capita welfare benefits tend to be somewhat higher in DC than the suburbs, and that DC's city work force is somewhat less productive, the burden of the poor on DC taxpayers is likely to be eight times higher than for the surrounding suburbs. There is no way for the core city to be tax-competitive with its suburbs under such conditions.
For instance, the most popularly accepted solution to all DC's current problems is simply to attract more residents back into the city, bribing them if necessary to return. Since we now have more poor people in the inner city than in these five neighboring jurisdictions put together, we would have to import over 3,000,000 residents to spread the poverty burden equally. Clearly, some other solution is necessary. Their cost must be shared with the suburbs (by some form of transfer payments), or perhaps even encouraging some of the poor to physically "follow the tax base" to the suburbs. In its starkest form, this equates to a grim bargain with the suburbs--"send us some of your tax receipts, or we must send you some of our highest tax consumers!"
David Rusk's new book, Inside Game, Outside Game paints an illuminating, if grim, picture of the changes that have taken place in the demography of poverty over the past half century--and probably due to some extent, to the increase in statistical completeness. Prior to the advent of Social Security, Medicare, and Medicaid, the majority of the poor were old men and old women. Today, the vast majority of the poor are single moms and their kids, and the vast majority of this cohort lives in concentrated urban poverty, in the large housing projects now considered by most urban practitioners to have been an unmitigated--if well-intentioned--disaster. It is also primarily black.
The impact of the single-parent family seems fundamental to the problem. Among married families, poverty afflicts 4% of whites, 11% of blacks, and 12% of Hispanics. among single moms, white poverty rises eightfold to 33%, but for blacks and Hispanics, it jumps to 56%. Only naturally, then, poverty among children is similar: 35% of white poor, and 46% of black and Hispanic poor.
But most troubling is the relative concentration of poor in urban ghettos--and its self-perpetuation through what Rusk calls the "poverty machine", and what NARPAC calls the 'poverty traps'. Of those fitting the official category of poor, 45% of whites live in the central cities, 67% of Hispanics, and 77% of blacks.`But among those urban poor, only 25% of whites live in high-poverty neighborhoods, while 50% of Hispanics, and 75% of blacks are clustered in substandard living conditions--where the poverty and all its trappings feeds on itself. To make matters far worse, it is not only the home environment which impacts on the kids, but their unsafe neighborhood environment, and their substandard school environment as well. And finally, as the limited-skill jobs also move away from the central city, then employment opportunities for the single moms decline in what is now referred to euphemistically as the "jobs/housing mismatch".
The absence of two potential breadwinners per household is clearly a critical factor in the continuation of poverty--and getting steadily worse. Across the US, there are now 5 white married couples per single mom; 3.1 Hispanic couples, and only 0.8 black couples per single mom. In fact, the DC metro area is noticeably more conservative than the national norms, with 6.4 white couples, 4.3 Hispanic couples, and 1.2 black couples per single mom. On the other hand, conditions in DC's poor tracts are worse than the national average: nationwide, the mean is 0.43 couples per single parent in poor tracts compared to 3.5 in the rest of the metro area, while in DC there are only 0.17 couples per single mom, compared to 3.7 elsewhere in the metro area. In a period of roughly ten years, the number of census-designated "poverty tracts" nationwide has risen from some 3,250 to 5,000. All of the high-poverty neighborhoods in the DC metro area are within the central city, and most of them have until recently contained outmoded housing projects--which are now being almost totally renovated.
Government efforts to "disperse" these pockets of poverty are related to the availability of de-centralized public, subsidized, and affordable housing--and the willingness of the poor to relocate. The Section 8 voucher system was intended to encourage poor households to relocate--initially to mixed housing neighborhoods within a given jurisdiction, and more recently, across jurisdictions. It has long been noted that poor blacks tend to gather in public housing, while whites tend to accept the Section 8 vouchers and disperse to low poverty areas and beyond. At present in the DC metro area, however, there are many more vouchers issued than there are affordable homes available. Several suburban counties claim to be working on that problem.
Statistically, whites are also substantially more likely to be home owners rather than renters. But the implications of these sorting processes may not be so much skin color or extent of poverty as the absence of a second parent. Marital status may well influence the preference to hang together or to disperse, and whether to want a landlord or a mortgage. In fact, NARPAC suspects (without statistical support) that it is also more difficult for single moms to establish a basic credit rating. If so, then solving the challenge of relocation might better focus on how to increase the dispersability of poor single moms and their kids.
NARPAC, Inc. has virtually no expertise in divining the preferences of very poor, generally unemployed, single black moms living in crowded, unhealthy conditions in a discouraging--and generally threatening--environment. But the idea of striking out on their own with their fatherless kids into some alien middle class white suburb cannot possibly be very appealing--particularly without the means to afford a car. One wonders if there isn't some relatively unexplored middle ground that might allow two or three such incomplete households to band together into some ersatz extended family group that could work together to improve their collective lot. A socioeconomic entity of three adults and six kids (particularly if a grandparent could be added as well) could easily be far more successful in working its way out of poverty--and poverty housing--than one single mom, very likely a high school drop- out, and two deprived kids all alone. It is not clear to NARPAC that the various laws, codes, and rules of welfare, unemployment, and housing would permit such banding together for survival. It might be worth exploring.
Alerted by a listing in the Public Education Network Weekly Newsblast, NARPAC obtained a copy of a recent telephone survey of some 1952 American adults conducted during January and February of 2001 by the Kaiser Family Foundation, Harvard's Kennedy School of Gov't, and National Public Radio. 294 respondents were below the poverty line, 613 between 100% and 200% of the poverty level, and 1045 above the 200% dividing line. It did not, of course, include households without telephones. Among the more interesting results are the following:
o Only about one in ten Americans even within this sample identifies povety and welfare as one of the top two issues the federal government should address. When asked to consider poverty, however, a majority then acknowledge that poverty is a "big problem".
o 64% of Americans believe that the $17,029 poverty threshold for a family of four is too low and should be raised to $20,000, while 42% think it should be at $25,000.
o Not surprisingly, there are somewhat different views of the primary causes of poverty, but they are not very different, and the more important (to NARPAC, at least) are those expressed by the poor themselves, for whom drug abuse, medical bills, inadequate bills, and "too many one-parent families". Somewhat less important (as shown in the bar charts below) are the decline in morality, lack of motivation, a shortage of jobs, poor public schools, the welfare system, and at the bottom, too many immigrants.
o In addition, the general comments on poverty show that a surprising 35% of the poor believe they have "easy lives", and almost 40% believe "the poor really don't need help". Half believe that Welfare hasn't helped much (!), and 60% acknowledge that jobs are available, but very low quality.
o Equally interesting is the significant difference between Republicans and Democrats as to what causes poverty (lack of self-help or circumstances) whether the government can in fact eliminate poverty, and the impact of government benefits, as illustrated below:
Census data are now available to illustrate the composition of all American families below the poverty line from 1980 to 2000, broken out by race and by those with kids above and below 18 years old, and by single heads of household (HoH) female and male and married couple households. The chart below shows the total numbers of households in each category for non- Hispanic Whites, Blacks, and Hispanics.
There are interesting trends for each race:
o Whites make up the largest group of disadvantaged households, even though they are a considerably smaller share of total while households (5.3% v 19% for Blacks and Hispanics);
o only the Hispanic category appears not yet to have peaked, whereas there are significant declines for Whites and Blacks, in part presumably due to the improved economy, and in part due to the federal "welfare-to-work" program;
o a small majority of White and Hispanic households in poverty are headed by married couples (57% and 52%) compared to only 15% of Blacks, while 69% of black households in poverty with kids under 18 are headed by a lone female, compared to 35% for Whites, and 38% for Hispanics;
o Blacks and their kids in particular--are substantially disadvantaged by their failure to marry, improve their household income, and strengthen the guidance for their kids;
o But perhaps most important, 59% of all families below the poverty line have only a single adult as head of household 51% a female, 8% a male and almost half of those families (41%) are black. Poor white single moms average 2.25 kids, and poor black single moms average 2.75 kids.
o Other trends in poverty across the US over the past 20 years are illustrated below.
In 1980, there were 29,272,000 Americans below the poverty line. By 1995, that number had grown to 36,424,000. Then as the economy grew at an unprecedented pace, the total dropped back in the 2000 Census to 31,139,000. Here are the significant changes in the demographics of the poor over this relatively short 20-year period., as summarized on the chart set below.
Perhaps the greatest change for the poor is where they live. Rural poverty has dropped by almost 4.5 million, from 38% to 22% of the national total. Inner city poor have increased by 2.3 million, but in 1995 the increase had peaked at more than 5.5 million over 1980. Cities were home to 36% of our poor in 1980, 44% in 1995, and back down to 41% in 2000. In between, the biggest growth has been in suburban poverty, rising from 25% to 36% of the US total.
The share of the working poor has changed relatively little, from 26% to 27%. 11
million were too young in 1980 and 2000, though the number grew to 14 million in 1995. 38%
did not work at all in 1980, and 39% in 2000. The full-time working poor rose from 1.6 million
(5.6%) in 1980 to 2.4 million in 1995 and 2000 (7.8%), while part time workers numbered 6
million in 1980 (20%) and 2000, but had grown by 1 million in 1985-95.
Family composition has shifted significantly towards non-family poor, growing from 23% to almost 30% ( over 8 million) of the total in 2000. Well over 9 million people below the poverty line were in families headed by a lone female in both 1980 and 2000, though the total peaked 3.6 million higher in 1995. Over 10 million were in families with two adults at both ends, with a much smaller growth in between. (Note that these numbers are for total individuals in poverty, not households. By the latter count, more than 50% of all households in poverty are held together by lone females.)
Social researchers seem obliged to keep repeating that poverty per se does not produce kids who are unable to learn. On the other hand, they continue to gather essentially irrefutable evidence that the poorer the household, the higher the likelihood that circumstances will exist which will "harm children's development, and contribute to problem behaviors, failure in school, and poor mental health". Such conclusions reinforce NARPAC's longstanding contention that many of the things that produce poor school scores are outside the dominion of the schools to fix. Dispersing poverty so that its kids do not overwhelm inner city schools is much more important in the long run that pumping more funding and more teachers into the schools themselves.
A series of three papers has recently been published by the "New Federalism" program "to assess changing social policies" within the Urban Institute, that summarizes much of these data, much of it gathered from the recent National Survey of America's Families (NSAF). These brief papers are available on the Urban Institute's website and is summarized here.
The researchers have identified three separate categories of problems that impact on kids'--and parents'--"well-being", and provide separate statistics on each. The three areas are "sociodemographic risk", the basic household statistics; "turbulence", the dynamics of churning households; and "family stresses" which create additional mental burdens for the family unit. Each is discussed separately below, although the problems seem very difficult to isolate from each other:
Four conditions comprise this risk index, any three of which qualify the kid as "high risk". They are listed below along with the percent of the total 13-state NSAF sample that suffer that condition, followed by the percent of that group considered high risk:
o 20.4% live in poverty, and 28.6% of them are high risk;
o 18% exhibited prominent behavioral problems compared to 6% of other kids;
o 25% exhibited prominent behavioral problems compared to 7
% of others;
Six parameters were used to measure family turbulence:
o moving from one state to another;
o 13% of kids in families below the federal poverty level (FPL) experienced turbulence
compared to of 3% with incomes three times the FPL;
o 22% had low "school engagement" vs. 15%;
o 40% had low "school engagement" vs. 25%;
o unable to pay mortgage, rent, utilities some time in past year;
o 49% of kids with parents without high school degrees lived under stress, vs 7% with
one college degree;
o 31% of all school age kids in stressful situations exhibited low levels of "school
engagement", compared to 17% of others;
No attempt is made to fold together these three separate problem areas. In fact, it seems unlikely that it is practical to completely separate them. But there can be no question that all of these problems tend to afflict households near the bottom of the economic spectrum, and that many of the kids reared in these households are not only going to be substantially more difficult to educate, they are likely to impact the ability of the more fortunate kids around them. Surely the cures for these problems do not lie within the purview of the nation's public school systems.
There are any number of "traps" by which the nation's needy are caught in a cycle of disadvantage which can only make their situation worse. One simple example is when stores move away from a depressed area for lack of business, forcing those least capable to travel further for their necessities, and often to pay more for them. This section hopes to explore some of these traps.
o Automobile Insurance
Any number of circumstances combine to make the urban poor even poorer. For instance, the shift of lower paying, lower skilled jobs to the suburbs is one case in point. This trend is aggravated by the fact that these suburban jobs (frequently paying over 20% more than their in-city equivalents) are seldom concentrated near public transportation nodes. A GAO study of welfare administrators indicated that 75% of welfare recipients who wanted to work but had failed to find jobs gave 'lack of transportation' as a major (44%), or moderate (31%) reason. An associated factor is that urban car insurance premiums often become prohibitively high--causing both resentment and an urge to "get even".
A 1998 report by Congress's Joint Economic Committee--at the behest of Representative Jim Moran, among others, who had the DC area in mind--indicates that among adult recipients of AFDC (Aid to Families with Dependent Children), those who owned cars were 12% more likely to have a job of some sort, worked an average of 23 more hours per month, and had monthly earnings $152 higher. In some jurisdictions, cars are being offered to welfare recipients to help them get and keep jobs.
But the study also found that city insurance rates are from 25% to 33% higher than in the suburbs. Families earning less than half the poverty level spend an average of 31% of their income on car insurance premiums--if they carry it at all-- often paying over half the remaining value of the older car each year (to say nothing of the higher maintenance costs for those older cars).
All car insurance premiums are increasing (the national average was $774 in 1996) and rising faster than inflation rates for food, housing, and medical care. City premiums are higher for a variety of reasons (some controllable), several caused by policy holder avarice. First, the number of claims from city owners are far higher than their suburban counterparts--65% more claims for bodily injury and 38% more for property damage. More than 44% of these claims are against the city government, widely perceived to have the deepest of all pockets.
Second, even though urban accidents are generally less severe (at lower speeds), injury claims per accident are higher by from 69% (in Baltimore) to 350% (in Philadelphia). Moreover, injury settlements are higher, in part due to the perversities in the current tort system--which also lead to widespread (and highly organized) fraud and 'buildup' ("ghost riders", etc.) accounting for as much as 35% of all claims paid.
Transaction and litigation costs are higher in cities, as are the dubious related claims for "pain and suffering" (often included to help pay the legal fees). Only 15% of bodily injury insurance costs actually go to medical bills and lost wages: lawyers fees, "pain and suffering" settlements, and fraudulent claims take 58%. Finally, insurance premiums are surely higher in cities to compensate for uninsured motorists--many of whom may be striving to get out of their "poverty trap".
The Joint Economic Committee's study estimated the advantages of adopting "Auto Choice" as a national approach to auto insurance, involving primarily a "no fault and no pain" approach which it likened to "health insurance for auto owners". Auto Choice would lower premiums about 24% for all Americans, and return over $35 billion yearly to consumers' pockets. It would not in any way involve cost- sharing between city and country dwellers, but would lower urban premiums much more by eliminating some of the higher city cost factors.
NARPAC, Inc. believes that a somewhat less ambitious approach might benefit metro areas with high levels of inner city poverty. For instance, a special regionwide insurance plan could be developed solely for those below, say, twice the poverty level, and administered by existing insurance companies. Its purpose would be simply to facilitate car ownership for the poor by assuring adequate compensation against real losses, but no windfalls for imagined grievances. Such a program would be similar to the availability of low-rate mortgages from Fannie Mae. Alternatively, it might become the equivalent of "Medicaid for needy car owners" (Medicar?). In fact, the program might be combined with some method of providing low cost auto financing for used, reclaimed, or donated cars.
In short, if the suburbs are unwilling to share the major costs of central city poverty, or provide enough affordable housing to draw the poor to the job opportunities, then perhaps they could at least help level the playing field in assuring transportation from welfare to work.
o Rent Controls
Rent controls--often argued to benefit most the poor--are another form of poverty trap in that rents can be adjusted upwards when a tenant leaves. Hence the advantages of rent control only last as long as the tenants stay in their current, generally squalid accommodations. This topic is discussed in greater detail elsewhere.
It becomes abundantly clear that the costs of running a city are intimately related to the income mix of its residents. This has been shown in a recent NARPAC analysis concerning the varying 'productivity' of people. It delineates the significant differences between the high income taxpayers and the low income tax consumers. These productivity calculations estimated the level of services (e.g., education, police) required by people of different income levels. Hence the per capita costs of government can be estimated "theoretically" by putting together different income mix populations and calculating what the per capita average will be for the city.
In practice, of course, it is certainly not clear that DC's city service costs have changed very quickly with population changes. DC School costs declined somewhat with a smaller student body, for instance, but police costs have not declined with the current drop in population. Nevertheless, this exercise in perfectly responsive city costing is useful to illustrate the dangers of using various indicators inappropriately. For instance, NARPAC itself tends to use the total per capita cost of city government as an indicator of Inefficiency in city management even though it is equally directly related to the city's income demographics.
Five simple options are tabulated below, to demonstrate how much the per capita costs of city government will change with different kinds of incremental population changes (50,000). In the first option below, the city's middle income population drops from its current level of 300,000 in 50,000 increments, while the 50,000 high income and 150,000 poverty-level people stay constant (along with the 30,000 students, patients, and other non-household residents). The net result is that the per capita cost of government increases from its current figure of $8742 with a population of 530,000 to a level of $11,222 with a population of only 380,000:
Option 1: Reduce the Number of Middle/Lower Income Residents
This clearly shows the increasing dangers of losing a city's middle income population. Option 2 then takes this condition in the opposite direction, adding 150,000 middle income residents and bringing the city's population back up to 680,000 people, again with a constant number of rich and poor. In this case, however, the decrease in per capita costs becomes less pronounced, dropping only from $8742 to $7356 with a 50% increase in middle class residents:
(Case 1 = now)
Option 3 takes the more obvious step, and decreases the number of DC residents below the poverty level in 50,000 increments until there are none left. This same theoretical calculation shows that the per capita costs of city would drop from $8742 to $3782. There are in fact some smaller cities where the costs drop to near this level:
Option 3: Decrease the Number of Poor in DC
Option 4 again does the reverse, adding 150,000 to the numbers of poor in increments of 50,000. Here as in Option 1, the result is predictable: per capita costs are again increasing to approach those of the poor alone. Hence a city with 350,000 of its 680,000 residents poor will have a per capita government cost of $11,514--about the same as a city of 380,000 with only 150,000 middle incomers:
Option 4: Increase the Number of Poverty-Level Residents
The next option keeps a constant population level, and trades off poor for middle income residents. Clearly, this drops the per capita costs even faster than in Option 3, because the middle income base is being expanded simultaneously. Hence a city of 530,000 with no one below the poverty level might have a per capita cost of $3407:
(Case 1 = now)
But equally important here is to note how very much the per capita cost indicator changes within a city whose government is considered to be equally "efficient" in each case. City costs are clearly related as much to who the government is serving as to how efficiently it is operating.
In the final option below, the poor are shared with a jurisdiction (or jurisdictions) almost four times as populous, but with a far smaller share of people on welfare. In essence 100,000 poor are "traded" from the smaller entity (Case 1-2) to the larger entity (Cases 4-3). The "before" status is represented by Cases 1 and 4, and the "after" status by adjacent Cases 2 and 3:
(Case 1 = now)
As shown in Option 5, if DC could replace 100K of its poor with the same number of middle income residents, per capita government costs would drop from $8742 to $5186 (Case 1 to 2, here). On the other hand, a much larger jurisdiction of 1980K persons (the suburbs, collectively, say) would see an increase in per capita government costs from $3622 (Case 4) to $4599 (Case 3) if their poor increased from 100K to 200K, while their middle income cohort dropped from 1750K to 1650K. In short, the larger suburbs would experience a 25% increase in per capita government costs, while the smaller central city would derive a 41% reduction. The suburban costs would still be somewhat less than those for the center, but the share of the population below the poverty level would be more closely matched throughout the region. Such 'poverty- sharing' must become a realistic national objective, not just for the national capital metro area.
One of the basic reasons that inner cities with large black populations are often in financial troubles can readily be traced to the stark and troublesome difference in earning power between the two races. The problem is essentially three-fold. Readily available Census Bureau statistics show the relative numbers of earners per household, and their earnings. As shown on the two charts below, there are significantly more black households with either no earner or only one earner and thus far fewer two-earner or three-earner household (upper chart). To make matters more stark, the household earnings for whites in each category is substantially higher than for blacks (lower chart). It is interesting to note, however, that this gap closes (percentage-wise) at the higher household income levels:
While blacks have made some significant strides in the past ten years, they appear many years away from closing the gap completely. This is shown on the aggregate chart below comparing the total number of black and white households, and the change in both median and mean income. The white advantage remains very substantial. The significance of the difference between "median"' and "mean" (average) is demonstrated elsewhere in a NARPAC tutorial, but in essence, the lower the median relative to the mean, the bigger the gap between the number of poor and the number of rich.
Finally, demographics also play an inescapable role in this: Over half of typical city costs are spent on kids' valid health and education needs. Since more black households have kids--and more kids, at that--more demands are placed on the public sector. To make matters worse, fewer blacks marry than in any other American racial group, and this is reflected by their unusually high share of one-earner, low-earning households.
The combination of these four factors (number of earners per household; earnings per earner; public services per household by income level, and kids per adult) can be demonstrated in a blunt and admittedly hypothetical way. If DC had an all-black population of 284,000 households exactly matching the national average in demographic and economic proportions, it would generate a total Household Taxable Income (HTI) of $6 billion dollars, providing revenues of about $1.5 billion but requiring government expenditures approaching $3.3 billion. If, at the other extreme, DC was the perfectly typical all-white American city, it would require only about $2.2 billion in city services, far less than the $4.3 billion in revenues derived from a citywide HTI of $19 billion. It is an inescapable fact that the burden on that share of the resident population paying taxes will be much higher in a predominantly black city than in a predominantly white city. It may also be financially unstable, since no city's demographic mix can be ontrolled. This is not a social statement or criticism, it is an statistical fact of economic life: city finances should not depend primarily on variable residential futures The summary numbers are displayed below. For a more detailed display, please turn to NARPAC's new analysis of household productivity.
From NARPAC's analytical point of view, the very substantial increase in out of wedlock births is not worrisome from ethical or moral standpoints, but from economic and educational ones. Kids raised by a single parent are far more likely to live below the poverty line, and far more likely therefore, to do poorly in school, drop out sooner, and get trapped in a life of crime and correction. In 1960, some 5% of all births came from unwed parents. By 2000, that share has risen to a full 33% some 1,264,000 kids born with a less secure future in one of the world's richest countries, 793,000 of them white and 415,000 of them black.
As shown on the accompanying upper chart, American births declined from their 1960 level into the seventies, rose again to their 1960 rate in 1990, and fell again thereafter. White births of all kinds have remained remarkably stable at 80%, and black births at about 17%, but births to unwed mothers have risen almost sevenfold from 5% to 33%. Black out of wedlock births have grown more than threefold from 22% to a stunning 69%, while white o-o-w births have risen thirteenfold from 2% to 26%.
As shown on the lower chart , the black and white share of total o-o-w births have almost exactly reversed: black babies have dropped from 63% to 33% of the total, while white babies have risen from 37% to 63% of the total. In real numbers, then, since 1960, at least 27,530,000 "illegitimate" kids have been born in the US: 14,000,000 or so white, and 12,000,000 or so black. And surely there is a more than casual relationship between these figures and the fact that in 1960 there were some 1,373,000 single-parent white households with kids in poverty, and some 1,426,000 single-parent black households with kids in poverty. The vast majority of each is headed by a female, and with kids below the age of 18. (See families in poverty above.)
Sustained economic growth in the Washington metro area has finally led the states of Virginia and Maryland to confront the unusual issue that their revenues are growing faster than their planned expenditures. Both state legislatures are now addressing whether to spend the surplus or offer tax reductions to their constituents. These state surpluses for FY99 are very substantial indeed: almost $900 million for Virginia, and about $250 million for Maryland, for a total of $1.15 billion. DC also expects a surplus in the neighborhood of $400M, but its per capita tax rate remains very much higher than their neighbors'.
Some of NARPAC, Inc.'s earliest analytical work compared the relative per capita tax burden between DC and its two neighboring states. It was clear several years ago that whereas DC's taxes are more than twice (210% in '95) the national average, the burden on Virginia and Maryland were somewhat less than the national average (7% in '95). It provides just one more excuse--probably secondary to high crime and poor public schools--for taxpayers to move to the suburbs.
In Carol O'Cleireacain's seminal book "The Orphaned Capital" exposing many of the financial problems with DC's budget, she estimates in 1994 dollars, the income taxes lost to DC by taxing workers where they live rather than where they work. Of course, more than 20% of DC residents also work in the suburbs, so there is a "reciprocal"--though by no means balanced--problem. For 1994, she estimated that Virginia gained $366M in income taxes from residents working in DC, and Maryland, $619M. At the same time, DC benefited from $144M in income taxes from residents working in the suburbs.
Scaling up O'Cleireacain's estimates for the five intervening years NARPAC estimates that in 1999, Virginia and Maryland will enjoy $1.05 billion in revenues from income taxes on wages earned in DC. If those taxes went into DC's coffers, the overall per capita tax rate on DC residents would be lowered by 39%, and DC would be well on the way to becoming a competitive place to live, at least tax- wise.
Farfetched as it may seem, Maryland and Virginia's budget surpluses in 1999 will almost exactly match DC's increased revenues if income taxes were collected by place of work rather than place of residence. Before these two very well run states decide to lower their own tax rates, perhaps they should look again at the "free ride" they are getting at the District's expense. In addition to increased sharing for Metro and the region's transportation infrastructure, Maryland and Virgina should consider underwriting some of DC's common metro-area problems associated with the high costs of poverty, which threaten the entire region.
NARPAC estimates that the states could well together pick up $400M in current DC expenses by adopting equal sharing of costs for public health care ($180M), public safety ($150M), welfare administration ($35M) and public higher education ($35). NARPAC has sent open letters to the governors of each state suggesting such a "Payment in Lieu of (Commuter) Taxes (PILOT). Using O'Cleireacain's distribution of income earned by state residents in DC, Maryland would have to pick up $250M, Virginia $150M.
1998's Tax Revision Commission recommernded that the Federal Government provide such a PILOT to DC as a new form of federal payment(to compensate for lost real estate taxes on government property), as well as some form of gross receipts tax on businesses. This latter proposal is offered as an alternative to the commuter tax which cannot be levied on the states by DC because of their Congressional influence on the DC oversight subcommittees.
The Small Business Survival Committee has developed what amounts to a small business "misery index" which sums 11 major government-imposed or government-related costs impacting small businesses and entrepreneurs, from personal income tax to workers' compensation costs, for each state and the DC. The values for Maryland, Virginia, and the District are compared on the table below:
By these measures, DC ranks dead last, while Maryland ranks 20th among the states, and Virginia 18th. The average index for all 50 states is 36.40, while the median (25th) state (Massachusetts) score is 37.48. DC's score of 63.55 is well over 80% worse than its neighboring states. The leading states (not necessarily known for their entrepreneurial magnetism) are South Dakota at 17.82 and Wyoming at 18.55, and the worst states are Minnesota at 46.25, and Hawaii at 58.5. It might also be noted that California, New York, New Jersey, and Ohio rank from 42nd to 46th.
While not intending to detract from SBSC's worthwhile efforts, NARPAC, Inc. notes that DC is heavily penalized in this scoring by its crime rate (which is "normally" more than twice as high for cities as for states overall--see NARPAC statistics under the major issue topic of safety and justice), and an electric utilities tax which seems to be in a class by itself. Eliminating those two possibly anomalous critera still leaves DC at or near the back of the pack and well over 50% worse off than its immediate neighbors.
There are, of course, many other factors which contribute to a flourishing entrepreneurial environment, from cost of living and local markets, to accessible recreational facilities and school test scores. Nevertheless, there seems to be little reason for DC to resign itself to non-competitive tax scores, particularly relative to their most proximate competing jurisdictions.
This page was updated on Mar 5, 2002
© copyright 2007 NARPAC, Inc. All rights reserved