Historical Overview of Federal Spending: Minor Categories
As shown in my previous post, a minority of federal activities account for the vast majority of federal spending. These are National Defense and Veteran’s Benefits, Social Security, Medicaid and Medicare, and interest on the national debt. The national election should really be about those issues, and federal tax revenues. Yet I’m willing to bet that if your local incumbent member of Congress even bothers to campaign, rather than having secured a guaranteed re-election by being in a safe district and/or keeping challengers off the ballot, he or she will probably talk about anything and everything else.
In the line chart of federal spending that is part of the spreadsheet that is linked here, one can see that the dashed line – “everything else” – fell from 6.6% of GDP in FY 1979 under President Carter to 4.9% of GDP in FY 1995 under President Clinton, with the greatest decreases during the Reagan years. Post Reagan, the pattern had been that spending on everything else fell as a percent of GDP when the economy was strong and rose as a percent of GDP when the economy was weak – implying that it pretty much stayed the same as GDP did better and worse. In FY 2011, however, federal spending on “everything else” jumped to 6.8% of GDP, the highest level since the Carter Administration. To understand why, print out the spreadsheet chart on “Minor Spending Categories,” and read on.
For the purposes of the chart, I have subdivided “everything else” into three categories.
“Spending for the Needy” includes welfare (TANF), SSI (income support for the disabled), the Earned Income Tax Credit for the working poor (EITC), food and nutrition assistance (such as food stamps and school lunch), housing assistance (such as federal subsidies for public housing and Section 8 housing assistance to private landlords), funding for social services, federal unemployment insurance payments, disaster assistance, and farm income stabilization payments. With the exception of the EITC, which is structured as a reversible tax break, food stamps and farm income stabilization payments, most of this money is funded through state and local governments rather than being spent by the federal government directly.
With all the anti-welfare rhetoric of the 1980s and 1990s, I had expected that federal spending on the needy would have fallen as a share of the economy. Instead, total federal spending on the needy tended to rise as a share of the economy if recessions and fall in more prosperous years, as one might expect and the “Minor Categories” chart shows. Looking at the roughly comparable economic years, total federal spending on the needy equaled 1.9% of GDP in FY 1979, 1.8% of GDP in FY 1983, 1.9% of GDP in FY 1993, 2.0% of GDP in FY 1995, and 1.8% of GDP in FY 2004. The difference between Republican and Democratic administrations with regard to the needy, it turns out, was about one tenth of one percent of GDP.
While total spending was steady, however, the type of spending shifted considerably from the pre-Reagan years through the years of President Bush II.
President Reagan wiped out the Comprehensive Employment and Training Administration (CETA) program, under which those who could not get private sector jobs were put on the public payroll for make-work projects and training. Federal spending on training and employment fell from 0.42% of GDP in FY 1979 to 0.15% of GDP in FY 1983, and had drifted down further to 0.7% of GDP in FY 2004. Reagan also cut federal funding for social workers and services like foster care and assistance to the homeless, with a cut from 0.26% of GDP in FY 1979 to 0.16% of GDP in FY 1983, where federal spending on Social Services more or less leveled off.
From FY 1983 to FY 1995, federal spending on farm income stabilization payments fell from 0.60% of GDP to 0.09% of GDP, through some combination of program changes and higher food prices. And after the “welfare reform” of 1996, the cost of cash welfare payments for those who do not work and yet do not qualify for federal disability payments was either cut or shifted to state and local governments, as federal spending in this category fell from 0.23% of GDP that year to just 0.14% of GDP in FY 2011.
At the same time, federal support for the working poor increased with bi-partisan support, primarily through the Earned Income Tax Credit (EITC). Federal spending under this program increased from 0.03% of GDP in FY 1983, under Reagan, to 0.28% of GDP in FY 2004, under Bush II.
Spending on Housing Assistance, SSI, and food stamps was reduced as a share of GDP under the second President Bush, although the unusually mild early 2000s recession rather than policy changes may have been the reason. From FY 1995 under Clinton to FY 2004 under Bush, spending in these categories fell from 0.34% to 0.26% of GDP, 0.32% to 0.26% of GDP, and 0.51% to 0.39% of GDP, even as the Bush administration increased spending on just about everything except the poor.
It should be noted that most federal subsidy money for housing doesn’t go to support for poor renters, it goes to tax breaks for the large homes of homeowners affluent enough to find it worthwhile to itemize their deductions. The cost of the deduction for mortgage interest totaled $88.7 billion in FY 2011, when the cost of all housing assistance programs was just $59.2 billion. And that doesn’t even include all the contingent liabilities the federal government has taken on due to the nationalization of Fannie Mae and Freddie Mac and the refinancing of housing bubble era mortgages by the Federal Housing Administration. Basically, what the federal government has done for 60 years is spend lots of money to subsidize the development of the suburbs and newer cities in the Sunbelt to induce the better off the live there, while spending some money to turn older cities into a holding place for the poor.
With the Great Recession, federal spending on the needy has soared – from 2.23% of GDP in FY 2004 to 3.45% of GDP in FY 2011. Some of this spending is clearly cyclical, and in FY 2011, as in the comparable years of FY 2004 and FY 1995, federal spending on the needy was down from the previous year as the economy pulled out of recession. Federal spending on unemployment insurance, for example, totaled 0.89% of GDP in FY 2011, the highest since the same level was reached in FY 1983 in the aftermath of the worst recession until this one. This category of expenditure will clearly fall as the economy recovers and extended federal unemployment is cut back.
But some of the increase in spending may not be cyclical. As noted, the earnings of typical U.S workers has been falling, making more of them who would have been too well off for the EITC and food stamps now eligible to their earnings supplemented. Food stamp spending jumped to 0.71% of GDP in FY 2011 from around 0.5% of GDP in prior comparable years. While EITC spending fell, that might be because fewer low- and moderate-wage workers had earned income in FY 2011. It has been reported that the new jobs being created in the wake of the Great Recession pay 40% less than those lost during the downturn. Even after finding a low-wage, part-time job more Americans may remain eligible for food stamps and become eligible for the EITC.
We are heading, in other words, for an economy where the best jobs available in much of the country start at $9.00 per hour and top out at less than $20.00 per hour. That would provide an absolute peak annual income of $41,600 for those times when full time, year-round work is available, and a more typical annual income of $15,000 to $25,000, supplemented (or not) by food stamps, the EITC, and perhaps Medicaid when illness strikes. Cutting eligibility for and benefits under these programs would make the working poor poorer, but it would not increase their wages.
For purposes of the line chart, I grouped federal spending on General Science, Space and Technology; Energy; Natural Resources, Agricultural Services, and the Environment; Transportation; Community and Regional Development; and Education under the title “Spending for the Future.” With the idea that spending in many of these categories presents a current investment with a future payoff.
With the exception of occasional recession years, spending in these categories has gone down, down, down even as the national debt has gone up, up, up. Looking at the comparable economic years, spending in these “investment” categories was 2.62% of GDP in FY 1979 under Carter, 2.12% in 1983 under Reagan, 1.73% in 1993 under Bush I, and 1.63% of GDP under Bush II. There was little increase during the Clinton years, and it is likely that federal “investment” spending under Carter had already fallen significantly from what had been during the “nation building” decades through the early 1970s. “Investment” spending jumped to 2.04% of GDP in FY 2011, but that was likely a temporary increase due to a federal stimulus program that is expiring. With federal politicians unwilling to raise taxes or slow the growth of spending on today’s seniors, additional reductions in federal investment in the future seem likely.
This pattern fits right into a general pattern of generational inequity in public policy, with the generations now in charge inheriting a vibrant country from those who had worked and sacrificed for them, and then cashing it in to satisfy their sense of entitlement. And those who have read my posts over the years know that is a story was itching to tell. But upon further review, things aren’t quite so bad, or at least they haven’t gotten that bad yet.
A substantial share of the drop is accounted for by the “community and regional development” category, federal spending on which fell from 0.35% of GDP in FY 1979 to 0.09% of GDP in FY 1993 before leveling off. This represents, among other things, the elimination of the Urban Development Action Grants (UDAG) program. But unless you think state and local governments have spent too little money taking property by eminent domain to subsidize new stadiums, convention centers, and office and factory buildings for corporations they seek to attract from one place to another, little has been lost and UDAG is little missed. In fact, state and local governments have continued to make such investments, when it made sense even though they themselves had to pay for it, and sometimes even when it didn’t make sense.
Similarly, federal investment in energy fell from 0.36% of GDP in FY 1979 under Carter to 0.26% of GDP in 1983 under Reagan and 0.06% of GDP in FY 1993 under Bush I, before staying at about that level. One might say that unwillingness to invest in energy is responsible for our ongoing fossil fuel dependence. Except that economic history has shown that all that is required to reduce our ongoing fossil fuel dependence is reliably high prices, which encourage conservation, alternative energy, and domestic fossil fuel production.
The federal government could have enacted a tax that rose as fossil fuel prices fell, and kept that incentive in place, but Generation Greed couldn’t be bothered. As for having the federal government hand out money to decide how to shift our energy use, it isn’t to be missed, since its choices are based on campaign contributions and folly. Under Carter, the money was used for the “synfuels” program. After that debacle, most federal energy subsidies have been hidden in the tax code or regulation, and have mostly gone to the oil and ethanol industries.
Federal spending on education fell from 0.48% of GDP under Carter to 0.41% of GDP under Reagan, and stayed at about that lower level until an increase to 0.53% of GDP under Bush II. This was only partially due to policy. The number of students, as the Baby Boomers left school and enrollment fell, and their children entered school and enrollment rose, had something to do with it. But state and local governments increased education spending hugely on their own accord in the 1990s and 2000s, until the recent recession forced cuts. So variations in the small federal component of public school finance have had a limited effect.
Federal spending on general science, space and technology has been varied between 0.19 and 0.26% of GDP over the years, and was 0.22% of GDP in FY 2011, slightly higher than at the start of the analysis in 1979. No real decline there.
There was a decrease in federal spending on infrastructure as a share of GDP over the years, a temporary jump in FY 2011 excepted. This is the possible example of older generations cashing in the past and selling out the future. Federal transportation spending fell from 0.71% of GDP in FY 1979 under Carter to 0.60% of GDP in FY 1983 under Reagan, before falling to 0.53% of GDP under Bush I, where it approximately remained under Clinton and Bush II. Spending on Natural Resources, Agricultural Services and the Environment fell from 0.52% of GDP in FY 1979 under Carter to 0.40% of GDP in FY 1983 under Reagan to 0.34% of GDP under Bush I, where it too leveled off. That category not only includes environmental regulation and agricultural extension services, but also includes water projects, sewer projects, environmental remediation (as for Superfund sites) and investment in our national parks. So the evidence is clear: the United States is disinvesting in its infrastructure as a result of federal policy.
Or is it? The federal government doesn’t spend much money at all in these categories DIRECTLY, and never has. It merely passes money on to state and local governments, which do the work. When I analyzed state and local government capital construction expenditures on infrastructure (not buildings) from FY 1972 to FY 2008, as can be seen in the spreadsheet attached to this post, I found there was a big drop from FY 1972, at the end of the nation-building era, to FY 1979, but not clear pattern of decrease afterward. State and local capital construction expenditures on infrastructure equaled 1.7% of GDD in FY 1979, fell no lower than 1.4% of GDP in the 1980s after leveling off at about that level, and was back up around 1.6% to 1.9% of GDP in the 2000s. All that is well below the 2.3% of GDP state and local governments spent on infrastructure construction in FY 1972, and more was probably spent in earlier years. Perhaps European countries have better infrastructure because they kept spending at that level or higher, and we did not. But that decision pre-dates the current political era.
So it appears that the post-1979 drop in federal funding for infrastructure was mostly offset by higher state and local funding. Now if that state and local funding came from rising debts, as at the MTA, then there is good reason to believe the future of our built environment – and our economic future – has been mortgaged. But that would show up as higher debt, which is a separate issue to be dealt with in the next post. Thus far, in other words, the generations in charge did not stop spending to maintain the infrastructure, they simply passed the bill to future generations and perhaps provided lousy value for the money spent.
We conclude our look at federal spending with the catchall category “Other.” The general trend in “Other” spending had been a decline in the Reagan years followed by a leveling off at around 1.0% of GDP, with some variation along the way.
“Other” includes the rather small amount of money spent on international affairs, including the even smaller amount spent on foreign aid. Looking at comparable years, spending on international affairs has been flat at about 0.22% of GDP aside from a jump to 0.34% of GDP in FY 1983 under Reagan, and to 0.37% of GDP in FY 2011 under Obama. An increase in international humanitarian assistance was indeed part of the total increase under President Obama. It would not surprise me if the devastating earthquake in Haiti in 2010 was responsible for that increase in aid.
Spending on the Administration of Justice, including the federal courts and the Justice Department (and the FBI), was 0.17% of GDP in FY 1979 under Carter and 0.22% of GDP in FY 1995 under Clinton – before jumping to 0.38% of GDP in FY 2004 under Bush II and 0.40% of GDP under Obama. It would not surprise me if 9/11 had something to do with this increase, as spending on federal law enforcement activities in straight dollars nearly tripled from FY 1998 to FY 2011. They certainly weren’t pursuing white-collar criminals during the Bush II administration.
General Government spending isn’t even worth tabulating separately, except that in FY 1979 the federal government had spent 0.33% of GDP on general revenue sharing assistance sent to state and local governments. That fell to 0.18% of GDP in FY 1983 under Reagan and virtually nothing ever since. Today, virtually all federal money passed on to state and local governments is for specific purposes. The near elimination of general aid accounts for the only real movement in the “other” category until FY 2009.
That year, “other” federal spending soared to the moon, nearly reaching 3.0% of GDP before retreating. Where did it go? The straight dollar data for FY 2009 shows a massive increase in spending on “other advancement of commerce,” to $170 billion. It was the Wall Street and auto bailouts, or some of them. This is another category where I believe the huge increase in spending under the late Bush II and Obama Administration was set in motion by the economy. While spending has temporarily fallen, and even gone negative as bailout money is repaid, I’m afraid we are far from the end of financial bailout spending.
Off the books, the federal government took on enormous contingent liabilities when it decided to “socialize the losses” and protect holders of Fannie Mae and Freddie Mac bonds, making them whole for 100 percent of their investments. With safety as if they had invested in U.S. Treasury bonds, but with a higher interest rate than the suckers who actually did invest in Treasuries. This decision was made by Hank Paulson in August 2008. As a result of that decision, and the decision to use the Federal Housing Administration to refinance private loans to get the banks out of trouble, just about every time someone defaults on their loans federal taxpayers (or, since the money is borrowed, future Americans) take the hit.
The mortgage and home equity loans that are defaulting were used to pay inflated prices for houses in the housing bubble, a loss for the buyers, often first time homeowners, but a windfall for the sellers, often older retirees who cashed out. Some of the mortgages were used for cash-out refinancings to buy new SUVs and similar items. Some of the mortgages led to more and bigger new McMansions being built and sold than would otherwise have been the case.
Over time, as more and more loans default, more and more federal tax dollars are going to be shifted to spending on the purchase of SUVs and the construction McMansions in the past. The decision on the future of federal investment during the next generation, therefore, was made by the private sector during the Bush II Administration. The United States has invested in SUVs and McMansions, not roads and transit and the electric grid and water supplies and sewer systems and national parks. And there is nothing, short of defaulting or partially defaulting, on the Fannie and Freddie bonds that can be done about it.
The acknowledged federal, state and local debts are on top of that. I’ll discuss them next and last.