Sold Out Futures:  A State-By-State Comparison of State and Local Government Debts, Past Infrastructure Investment, and Unfunded Pension Liabilities Through FY 2019

In two years of the COVID-19 pandemic, with society under stress, we have seen increasingly strident political fights over whose cultural attitudes and preferences should be imposed on others, who should get to contribute less to the community, and who should get to take out more.  In the shadows, however, is a bipartisan consensus as to who should be made worse off and be sacrificed the rest of their lives to pay for it all.  Ordinary people in later born generations, those who will be living in the United States in the future.   The pandemic has given politicians of all alleged views, and the interest groups that back them, an excuse to do, to an even greater extent, what they have done for 40 years.  Cash in the common future to address the perpetual “emergency” of the present.

So it was in Washington in 2020 when The Donald and the Republicans, having already sent the federal debt soaring to cut taxes for the rich and then ran a federal deficit equal to one-quarter of the U.S. economy.

And so it is in Washington today, where Biden in the Democrats claim their plans will be “paid for” – meaning the burden shifted to the future would only be as great as it was under Trump and the Republicans.

It is in this context that for the fifth time, I have reprised an analysis of state and local government finance data from the U.S. Census Bureau, for all states and for New York City and the Rest of New York State separately, with data over 49 years, to determine the extent to which each state’s future had been sold out due to state and local government debts, inadequate past infrastructure investment, and underfunded and retroactively enriched public employee pensions.   You’d think that the extent of disadvantage for the later-born, and who benefitted from creating it, would be the number one issue in every state election, and the number one topic of debate in the media.  Instead, it remains under Omerta, especially here in New York.  Shouted down under the comforting culture war issues that Generation Greed prefers.  So, although standing up for the later born and common future may amount to nothing more than standing on the beach shouting into a hurricane as a social tsunami heads for shore, over the past month I have updated the “Sold Out Future” analysis with data through FY 2019.  This post, a national summary and explanation of where the data comes from and how it was used, and the next three, will show what I found.


The Census Bureau data on state and local government finances may be found here,

With more specific data on public employee pensions here.

This dataset goes back to the late 1950s, with readily available data available for 1972 and each year from 1977 to 2019 (other than 2001 and 2003, when budget cuts cancelled the survey).  All state and local governments are surveyed every five years, but only state-level estimates are produced in the survey years in between.  Since the City of New York and the Port Authority are always in the survey, however, and are the only significant local governments within the city’s borders, data for each year may be produced for NYC — and the Rest of New York State by subtraction.  Given some effort to allocate the finances of the State of New York between those areas, and allocate the Port Authority between NYC and New Jersey.

The Census Bureau once produced spreadsheets with data for all data items for all years for each state, for state governments, local governments, and state and local governments combined.  The files were referred to as DAC-REX, and ended in 2008.  I have spent weekends over the past month (November 2021) appending data for additional years to those files in order to do this analysis.  What might be easily done by others has been difficult and time consuming for me, I have done it anyway.  Bear in mind that although there are tricks I use to avoid making errors, or find and correct them, no one has checked this information for me.  Having spent so much time already, I’ll re-use some language from the FY 2016 analysis, where appropriate, especially in this background post.

It isn’t just the dollar value of the hole each state and its local governments are in that matters.  It is also the total income of all those who live there, that could be used to get out of that hole.  As is generally the case with my analyses of state and local government finance data, the most critical measures in this analysis are based on percentages of state residents’ total personal income.

Data on the personal income of all residents of each state, and New York City and the Rest of New York State separately, was downloaded from the Bureau of Economic Analysis’ Local Area Personal Income series, located here.

That data is readily available starting in 1969.  Based on an analysis of the data through 2019, I found that New York City has recently had the greatest economic boom relative to the rest of the country since the 1920s.

So as you examine how “Sold Out” New York City’s future is relative to the Rest of New York State, the U.S. average, and other states, remember that you are looking at a best case scenario, with a huge number of young working adults packing into apartments at high rents, being paid less, receiving relatively few public services, and having their taxes go toward costs from the past.  Were that to end, and fewer taxpaying young exploitables were here to cover the costs of the past, NYC’s Sold Out Future total would be much worse.  

This post, however, will take a national focus, with charts showing the pre-pandemic trend for all state and local governments in the U.S. combined.   That will set the stage for the following posts, highlighting the worst off and best off states by each measure, and overall.  

Here is the spreadsheet with the entire database, some tables, and most of the charts, I plan to use in all the posts (although I generally end up adding charts once I start writing).

We begin with the level of state and local government bonded debt, the one aspect of selling out the future that isn’t “hidden,” and the one that has been getting better recently.  The following table shows national average debt and infrastructure trends as a percent of personal income.

State and local government debts equaled 17.2% of the total personal income of all U.S. residents in 1972, at the end of an era of extensive infrastructure, public school and higher education construction, with the latter developments needed to accommodate the youthful needs of the Baby Boomers.  The high inflation of the 1970s reduced the debt level to just 14.1% of personal income in 1981, when an era of financial irresponsibility took hold across the board in the U.S.  State and local government debts soared to 22.6% of U.S. residents’ personal income in 2010, but subsequently fell back to 17.2% in 2019.  That’s the same place it had been in 1972, but this time without the previous infrastructure investments.  Rising income, as the country recovered from recession, was a factor in the modest decrease in debt over a decade, but falling income in the next recession will increase the burden of that debt further.

While state and local government debt has been falling recently, moreover, that hasn’t been the case for total government debt as a whole.  Federal debt has soared since the start of the Great Recession, to keep the debt-burdened U.S. economy from collapse.

In fact, all categories of debt soared in the run-up to the Great Recession.  From then until 2019 household debt had only fallen because so many households defaulted on their mortgages.  And in 2020 all categories of debt soared more than before, with total U.S. debts for 2020 already in excess of the level of 2008, before the financial crisis and Great Recession.

Given the high level of debt in the economy, the only reason why U.S. federal, state and local governments are not yet experiencing drastic austerity of the kind often imposed on developing countries as part of IMF bailouts is the low level of interest rates.  

From 0.68% of U.S. residents’ personal income in 1972, the burden of interest on state and local government debts soared to a peak of 1.28% of personal income in 1987.   Falling interest rates, in part due to lower inflation, in part due to an influx of foreign investors buying up U.S. government debt with dollars gained from our persistent trade deficit, reduced that interest payment burden to 0.88% of personal income in 2007, despite soaring debts.  That freed up money for other things.  After a small increase during the recession, due to falling personal income that made the government interest payments more painful, the interest burden of state and local government debt fell to just 0.71% of personal income in 2019, the least since 1977.

Starting in 1982, the Census Bureau started tracking state and local government debt incurred for a “private purpose,” such as the industrial revenue bonds floated to subsidize commercial and industrial development to attract and retain businesses, separately from general government debt.

This “private purpose” debt accounted for 15.8% of total state and local government debt in 1982, but increased to 34.3% of total debt in 1990.   Borrowing money to pay for “corporate welfare” has become increasingly controversial over the years, and this may have affected political behavior.  Private purpose debt fell to just 17.6% of total state and local government debt in 2019, according to data reported to the U.S. Census Bureau.  

Or the controversy may have just led to misreporting.   The private purpose debt of the State of Alabama and its local governments, according to their reporting, fell from nearly $4.8 billion (44.3% of total debt) in 1991 to just $571 million (1.9% of total debt) in 2019.   Has Alabama stopped borrowing money for economic development projects to lure corporate investment, while paying off those debts from the past?  It’s possible.  But Baby Boomers have not been known for their veracity in any aspect of life, and it is also possible that Alabama politicians have done the same sorts of things with public finance data that New York politicians – and accountants, pension actuaries, bond raters and executive pay consultants – have done with other data.


One other use of public debt is investment in the infrastructure.  

My tabulation of past infrastructure investment includes construction expenditures (Census Bureau code F), but not other capital expenditures (code G).   Other capital expenditures include the acquisition of land, for new infrastructure, and equipment, such as new subway trains and buses.  I exclude land acquisition because that is only required in developing areas, whereas all areas need to either build new infrastructure or rebuild the infrastructure they already have.   I exclude equipment because subway trains and buses are like the private motor vehicles drivers buy themselves, whereas mass transit construction is like construction on the roads and bridges those drivers use.   This makes infrastructure investment more comparable across areas.

The tabulation includes transportation and environmental infrastructure (water, sewer, solid waste) while excluding construction in the Census Bureau’s “Parks” category.  That category includes sports stadiums and convention centers.  I wouldn’t say that such investments are unimportant – one of the purposes of cities from time immemorial has been as a setting for public religious and cultural gatherings.  But I do believe that U.S. state and local governments have over-invested in these types of facilities but underinvested in basic infrastructure, to our long-term detriment.

Similarly, with regard to capital construction in education, I only include the “other education” construction category and not “auxiliary enterprises” construction, because I want to exclude college sports stadia, dormitories and cafeterias.  Not because I’m against those things.  Because as in the case of professional sports stadia and convention centers, I believe the level of investment in them has been excessive relative to other things.  So excessive, in fact, that my compilation of the 2017 Census of Governments found that whereas tuition is covering more of the cost of actual state higher education than in the past, taxpayers are picking up more of the share of auxiliary enterprises, enterprises that once paid for themselves.

To calculate the extent to which each state had a “sold out future” with regard to infrastructure construction, I had to make a decision what the appropriate level of past infrastructure investment was, as a share of the total income of U.S. residents.   I had to create a benchmark to measure each year and state against.

There can be too much construction as well as too little, wasteful excess public investment as well as wasteful, excess private investment.  In the 1950s, 1960s and early 1970s, the federal government provided so much money for brand new state and local government infrastructure, mostly in the suburbs and Sunbelt, that perfectly good infrastructure the older cities of the Northeast and Midwest had paid for themselves were left to decay and were abandoned, even as those cities were taxed to pay for infrastructure elsewhere.   Transit systems, in particular, spiraled downward, along with the private railroads.  Instead of more infrastructure, the U.S. just ended up just building new places that devalued others.  It was only when that torrent of “free” federal money slowed, and some of it was shifted to reinvestment in older infrastructure, that some older cities such as New York began to recover.  

Toward the end of this era, in 1972, U.S. state and local governments collectively spent 1.63% of U.S. residents’ personal income on infrastructure construction, with investment in highways and streets predominant.  At the time, to quote the rock band Rainbow in the song “Can’t Happen Here,” there were concrete racetracks nationwide, juggernauts carving up the countryside, cars by the million on a one-way ride, using up the future.

Just remember, global warming is only the fourth negative consequence of U.S. dependence on fossil fuels in general, and imported oil in particular, after the environmental, economic and national security consequences known 50 years ago – and never addressed.  I guess they weren’t “worried about the future” after all!

One does not necessarily need to believe the estimates of required infrastructure investment produced by engineering and construction lobbyists.  Lobbyists for every type of public spending assert that far more taxpayer dollars should be spent on theirs, compared with today.  And with regard to infrastructure those estimates include upgrades to “modern standards” for infrastructure that is “functionally obsolete,” not just the maintenance and renewal of what we already have.

Continuing to invest the 1.63% of personal income might be what would have been required if Americans wanted the sort of infrastructure quality one finds in Europe and Asia, and were willing to sacrifice some of their personal standard of living to pay for it.   With, perhaps, more invested in transit and environmental infrastructure and less in roads than back in the 1950s and 1960s.  I wouldn’t be fair, however, to blame older generations for selling out our future for merely failing to keep up with other developed countries.   Rather, they are to be blamed for providing an infrastructure to their children inferior to what they inherited from their parents.

For that I use the 1.36% of personal income averaged from FY 1977 to 1981 as the benchmark.  I think that is a fair guesstimate of what would have been required to keep things from getting worse.  

There was another decrease in investment in state and local infrastructure construction expenditures as a percent of personal income after 1981.  From 1982 to 2019, U.S. state and local governments spent an average of just 1.08% U.S. of personal income on that investment.  This difference, small in any one year, adds up to more than $1.4 trillion in today’s money for all available years combined.  That is a hidden debt equal to 7.2% of all the personal income of everyone residing in the United States.  Both the dollar figure and the percent of income are worse than the last time I did this analysis, through FY 2016.

The visible state and local government financial debt in FY 2019, for comparison, was just over $3.2 trillion, or 17.2% of everyone’s personal income.

The drop off in transportation investment is small, but it adds up and has consequences over four decades.  U.S. state and local government transportation investment averaged 0.81% of U.S. residents’ personal income during the benchmark 1977 to 1981 period, but this fell to 0.75% of income on average during the 1982 to 2019 period.  There was a shift from highways to mass transit, but the decrease in highway investment was greater than the increase in transit investment.  And the increase in transit investment has apparently been insufficient to reverse the disinvestment in rail-based systems of the highway building era, leading to service meltdowns throughout the country – New York, Boston, and Washington.   Even “new” systems such as the DC Metro and the Bay Area’s BART have extensive components reaching the end of their useful lives at around age 50.  The Baby Boomers got their roads, but haven’t maintained them, and the Millennials are being denied the mass transit they seem to prefer.

Moreover, some of what is being classified as mass transit infrastructure is actually investment in private real estate development that does little or nothing for actual transportation.  New York City’s Penn South project, for example, uses transportation as an excuse to subsidize office development.  The Moynihan Station project is really just a shopping mall, as is the extremely costly Port Authority Trans Hudson station at the new World Trade Center.  These projects are much like the stadia and convention centers I have excluded from the “infrastructure investment” total.

In addition to that, there is the diminishing value received for the infrastructure investment that does take place, epitomized by the East Side Access project in New York City that took 25 years and nearly $12 billion to connect an existing tunnel to an existing station.

Lengthy delays have also affected New York’s East Side Access extension of the Long Island Rail Road, which is supposed to cut up to 40 minutes off commuter time on the last segment, from Queens to Grand Central Terminal, with up to 24 trains per hour at peak times.

Conceived more than a half century ago, with a construction contract awarded in 2006, that project was supposed to be completed by 2011. Early estimates put the cost at $2.2 billion, then $4.3 billion in 2006 and $6.4 billion in 2008. The Metropolitan Transportation Authority now envisions completion in December 2022 at a cost of $11.1 billion. Design changes, underground tunneling problems and coordination with other agencies were some of the factors in the delays and cost increases.

On the environmental side, one might have expected an increase in state and local government infrastructure construction expenditures as a percent of personal income, in the wake of the environmental movement of the 1970s.  Instead, after having averaged 0.38% of U.S. residents’ personal income from 1977 to 1981, state and local government construction expenditures on water supply, sewage and solid waste facilities and infrastructure averaged just 0.25% of personal income from 1982 to 2019.  

This is a surprise from a New York City perspective, where investments in such facilities, often in response to lawsuits and federal mandates, has crowded out transportation investments for 40 years.  But across America, apparently no one was going to force the politically powerful suburbs to replace their cheap cesspools with expensive sewer lines and sewage treatment facilities, to prevent environmental problems such as the algae blooms in the Long Island Sound – and the lead poisoned water mains of Flint Michigan.  

Going forward, however, it is possible that global warming will force a new wave of environmental investment, whether affordable or not.   In the past, most such investment, for flood control for example, has been undertaken by the federal Army Corp. of Engineers, not state and local governments, and is thus not included in this data.

Some point to a decrease in federal infrastructure investment as the culprit in our sold out infrastructure future.  But the data doesn’t show this.  From 1977 to 1981, federal aid to state and local governments in infrastructure categories averaged 0.42% of U.S. residents’ personal income.  From 1982 to 2019, the average per year was only slightly lower, at 0.40% of personal income.  During the weak economy years of 2002 to 2014, when personal incomes were lower, the average was 0.44% of personal income.  The problem apparently hasn’t been that state and local governments have been getting less in infrastructure funding from the federal government.   State and local governments have been spending more of their own resources on other things – like public employee pensions – instead, and/or cutting taxes.  And getting less value for money they do spend on infrastructure.

State and local government construction expenditures for actual education buildings tends to follow demographic waves, so we can’t really use one year or era as a benchmark.  Instead, the analysis assumes the average U.S. level of education building construction for 1982 to 2019 as a whole was appropriate, and compares each state with that. 

There was a high level of spending from through the 1950s, 1960s and early 1970s to accommodate the education needs of the large Baby Boom generation.  In FY 1972, at the end of that era, education construction expenditures were at 0.61% of personal income on average nationwide.  This fell to a low of 0.19% of personal income in 1984, with the smaller “Baby Bust” generation (Gen X) in school and plenty of building space left over from the Boomers.  Many schools were closed in the late 1970s, 1980s and early 1990s.  With the children of the Baby Boomers (aka the Baby Boom echo or Millennials) entering school in the 1990s, however, expenditures on education construction then soared to a peak of 0.65% of personal income in 2009 — just as the Millennial were exiting school.  

The question is will U.S. education buildings be maintained for the next wave of children and young adults, or allowed to deteriorate and be abandoned?   The 1982 to 2019 average level of spending was 0.41% of personal income.   The 2019 level was 0.41%.  

New York City is building new schools despite plunging enrollment as an excuse to keep politically powerful unionized teachers on the job in empty classrooms, even as those teachers fight to be allowed to work from home on Zoom.  And here, as well, one there is little value being received for the money spent.

In the Bronx, a 46,000-square-foot expansion of PS 33, which will add 388 seats, costs a reported $70 million. In Jersey City, a new 53,000-square-foot school that will serve 480 students, BelovED Charter High School, reportedly costs just $12.5 million.

The Big Apple construction bill is more than $1,500 per square foot — a price that would be too high for a “crazy-fancy private hospital,” let alone a school, one industry source told The Post.  The tab in the Garden State comes to less than $250 per square foot…

When presented with the Bronx school’s cost, one real-estate official, who declined to be named, said $1,500 per square foot was outrageous — and that the “only way” costs could run that high for a school is if a public authority was behind it.

One key factor in these high costs – retroactive increases in pension benefits, and pension underfunding, in the construction industry around the year 2000, leading to crises for multi-employer pension funds.  The construction industry and construction unions have cut deals with New York politicians to have inflated costs for public projects cover that bill, even as private projects go non-union, as I showed here.


Inadequate past state and local government infrastructure investment is a nebulous, hidden debt that older generations of Americans, through their politicians, have foisted on those who will follow them.   So are inadequately funded and retroactively enriched public employee pensions, as shown in the following table.

Financial advisors traditionally recommend that those about to retire accumulate enough in savings so that they can live off 4.0% of the initial asset accumulation in the first year, withdrawing a similar amount thereafter.  Although in this low inflation, low interest rate, low return environment, some believe this is too high. 

I also believe that state and local government public employee pension funds should have AT LEAST enough in assets that annual pension benefit payments equal just 4.0% of those assets.  Even though the actuaries, and city and state comptrollers and legislators that signed off on all the retroactive pension increases and pension underfunding in the past would probably deny this.  Given the situation they have left us in, do you believe them or me?

Benefit payments equal to 4.0% of assets, or perhaps a little more, may be an acceptable ratio for “mature” pension funds, where the ratio of retirees to current workers fully reflects the richness of the pension benefit, in years retired relative to years worked.  In fast growing localities and states, however, there will be relatively few retired public workers drawing benefits for work done during a past with a smaller public to serve.  And more active public workers on the job today providing for a larger current population, for whom retirement assets need to be accumulated.  In growth areas, therefore, benefit payments for that relatively small number of retired workers should be a much smaller percent of those public pension fund assets.  Thus a 4.0% benchmark for a state or nation seems reasonable.

For all U.S. state and local government pension funds combined, however, benefit payments only even approached the 4.0% standard during the late 1990s stock market bubble, which temporarily inflated pension assets.  Benefit payments to retired state and local government workers were at an average of 6.8% of pension fund assets in 1977 – a time when older central cities such as New York, fully built out and no longer growing quickly, were being crushed by public employee pension benefits that had been retroactively increased and underfunded to benefit generations cashing in and heading for the suburbs.  As I described in this popular post.

Benefit payments then fell to an average of 4.3% of pension fund assets in 2000.  The inflation of the 1970s reduced the real value of the pensions who had retired earlier, and then the falling inflation and interest rates of the 1980s and 1990s caused stock and bond prices – and pension assets — to soar, at least on paper.   The bubble further increased fictional pension fund asset values relative to benefit payments, even as actual cash investment returns – interest and dividends — decreased.  This was used as an excuse to cut taxpayer pension funding, increase public employee pension benefits, and cut pension contributions by the public employees themselves, all over the country.  Exactly repeating the deals that had led to fiscal disaster in New York and many other older cities decades before.

During the financial crisis, in 2009, state and local government pension fund benefit payments equaled 8.1% of pension fund assets.  Meaning there was, on average only enough in assets to pay for 12.3 years of benefits.  This despite the fact that public employees get to retire young compared with other Americans, and are living longer. 

The stock market has soared since 2009, and pension fund asset values have risen, again at least on paper.  But benefit payments still equaled 7.3% of pension fund assets in 2019.  That year, in order to meet the 4.0% standard, U.S. state and local government pension funds would have required more than $3.65 trillion in additional assets.  That is a hidden “pension debt” of 18.6% of all of the personal income of everyone in the United States.   While this is a back of the envelope estimate, it is similar to the current estimates of about $4 trillion in U.S. state and local government pension debt made by others.

Including a more detailed analysis by the Bureau of Economic Analysis for 2018, one that takes into account the fact that growing Sunbelt states ought to have even more money set aside than the 4.0% ratio would imply, to account for their larger number of future beneficiaries relative to current beneficiaries.  The BEA estimate put the total U.S. state and local government pension hole at $4.5 trillion, or 23.1% of the personal income of everyone in the country, as I discussed here.

Why did this happen?  For one thing the level of pension benefit payouts has soared as more and more public employees retire. 

State and local government pension benefit payments to retired public employees equaled just 5.2% of the wages and salaries of those still on the job in 1972.  That ratio has risen continually since, and benefit payments reached 31.3% of the wages and salaries of active workers in 2019.   

One wonders how high it will go.  If public workers were entitled to one year in retirement for each two years worked, and received an inflation-adjusted benefit equal to half their average pay, and the number of active public workers remained constant, as in a fully developed area with a mature pension fund, then the benefit payments to the retired would equal 25.0% of the wage and salaries of those still on the job.  That is a pretty rich retirement benefit compared with what most private sector workers receive.  And yet nationally total state and local government pension benefit payments have blown past ratio, with no end in sight.

So public employee pensions are really rich.   But that shouldn’t cause a crisis, because those benefits were supposed to be pre-funded while the public employees were on the job, and not cost another dime once they retire.  That is not what happened, however.  

Taxpayer contributions to the pensions of state and local government workers equaled 7.2% of the payroll of workers still on the job in FY 1972.  Those contributions increased to 9.8% of payroll in 1977 and peaked at 11.5% of payroll in 1984 and 1985.   

Then taxpayer pension contributions began to fall as a percent of payroll.  At the time, most private sector employers were eliminating defined benefit pensions for new employees, following the imposition of stricter funding standards under ERISA in 1974, but most state and local governments continued to offer them, since government funds were exempted from ERISA.

Governments promised pensions but did not fully fund them.  On average, U.S. taxpayer pension contributions to state and local government pension funds fell to 9.0% of payroll in 1993 and just 6.5% of payroll in 2002.  That 6.5% of payroll is nowhere near enough to pay for the one year of retirement for each year worked may public unionized employees have used their political clout to achieve, or even the shorter and less rich retirements many private sector workers expected from their 401Ks.  Worse, in many places public employee pension benefits were retroactively enriched at the same time, with no additional money contributed to pay for this.

Subsequently, taxpayer pension contributions had soared to 11.9% of payroll in 2012 and 16.6% of payroll in FY 2018, before dropping back to 16.2% of payroll in FY 2019.   That has been painful in the places where more and more taxpayer dollars are going to pensions, leaving less for other things.  In my estimation, however, it would take at least that much and more in total pension contributions to fund the kind of pensions most public workers have been getting, let alone the typically richer benefits for police officers and firefighters, even if there were no existing pension funding shortage to make up for.  So costs from the past continue to be shifted to the future.

State and local government employees themselves also contribute to their pensions, but at a far lower rate.  As I noted in this post…

One of the first public employee pension funds, that for New York City teachers, was established in 1917, and one of its principles was that the teachers and the city should each pay half the cost of funding teacher pensions.  

In 1972, however, state and local government contributions to their own pensions equaled just 4.3% of the wages and salaries.  Employee contributions stayed at or slightly below that level until the 4.3% recorded in 1995.  Since then, employee pension contributions have risen as a percent of payroll, to 4.6% in 2004 and 5.5% in 2019.    That is still far less than the 16.2% of payroll kicked in by taxpayers.

Bear in mind, however, that a significant share of public employees – including all teachers in states such as California, Illinois, and Connecticut, and all public employees in Nevada, are not in the Social Security system.  They will not be eligible to receive Social Security.  But neither they nor taxpayers are paying the 6.2% of pay up to around $142,800 under the payroll tax.   When allocating the pension blame between past taxpayers who didn’t pay enough in, and older and past public employees who got their pensions spiked and retroactively enriched, whether or not those public employees also get Social Security needs to be kept in mind.

Rising taxpayer pension contributions are an increasing burden on U.S. taxpayers, as they are paying more and more of their own incomes in taxes to fund them.  

Taxpayer contributions to state and local government pension funds equaled, on average, 0.57% of the personal income of all U.S. residents in 1972.  That increased to 0.75% in 1977 and stayed at about that level until 1987.   Then a generation or two decided to give themselves a gift, and taxpayer pension contributions fell to a low of 0.43% of personal income from 2000 to 2002.  This allowed more spending on public services without higher state and local taxes.  In some states it also provided political cover for pension benefit increases for public workers, who already had far richer benefits than other workers, most of whom get no retirement benefits at all.

Current and future taxpayers will pay the price.  Taxpayer contributions to state and local government pension funds jumped to 0.61% of total personal income in 2004, and have since soared to 0.91% of personal income in 2019.   There is likely a large share of Americans working in the private sector who are saving a lower share of their income for their own retirements than what they are paying in taxes for state and local government pensions.

Thus far, however, Americans aren’t really paying for the state and local government pension crisis in higher state and local government taxes.  At least on average, and not yet.  (There are widely divergent trends by state).

For the U.S. as a whole, state and local government tax collections have been in the vicinity of 10.0% of U.S. residents’ personal income for a long time.  Recessions, which reduce personal income, and asset price booms, which increase state income tax collections on capital gains, have temporarily increased that ratio from time to time, but the average state and local government tax burden hasn’t been above 10.7% of personal income since before 1977.  The 1982 to 2019 average is 10.1%.  The 2019 level was also 10.1%/

To the extent that some states and localities had increased their tax burden to pay for the soaring cost of pension benefits, others had cut taxes in the face of those soaring pension costs.

For the U.S. as a whole, though not for every state, rising taxpayer contributions to state and local government pension funds have been offset by lower total wages and salary payments to active state and local government workers.   State and local government wages and salaries fell from 7.9% of the personal income of all U.S. residents in 1972 to a low of 6.7% in 1984, leveled off or edged up for a time, but then fell to just 5.6% of personal income in 2019.  Mostly because there are fewer and fewer state and local government workers relative to the total population.

Not all of this is associated with service cuts.   Some of it has to do with the demographic waves discussed earlier.   Two of the three largest recipients of government spending in the United States are the traditionally female-dominated fields of education and health care (police, the military and similar are the third).   Whereas public spending on education has traditionally meant public spending on public employees – teachers in elementary and secondary schools and professors at state colleges and universities among others — public spending on health care mostly goes to private sector health care providers.  When you have a large generation – the Millennials — exiting school, and another large generation – the Baby Boomers – entering old age, one would expect government spending to shift from education to health care.  And, therefore, from public employee wages and salaries to Medicaid payments to the private sector.

In addition, some services have been privatized, in response to retroactive pension increases and low productivity in the public sector, and the inability to get fair value out of an entitled workforce.  Just as a wave of privatization followed the near bankruptcy of New York City in the 1970s.

Even so, there is no question that public services have been stressed, and public services constrained, in much of the country.   And the wages and benefits of newly hired public employees (like Millennials in general) have been reduced, in some places and cases drastically, compared with those hired earlier.  In Illinois, for example, new teachers not only do not get Social Security, but also contribute so much to their reduced pensions that they are in effect paying for nearly all of them.  With all the taxpayer contributions going to the richer, underfunded benefits of those hired earlier.  That union-friendly deal passed under a Democratic Governor, and with Democrats in control of both houses of the state legislature.  State and local government wages and salaries equaled just 5.4% of the personal income of Illinois residents in FY 2019, down from 6.2% in FY 2007.  Taxpayer pension contributions increased from 0.6% of personal income in FY 2007 to 1.75% of personal income in FY 2019.

In general, however, public employees hired earlier have given little if anything back.  And as a result, the average earnings (including the cost of benefits) per state and local government worker has increased over time when compared with the private sector workers who pay those bills.

But rather than debate the distribution of guilt between past taxpayers and public employee unions, some would prefer to convince later-born generations to blame Wall Street instead.    That is the scheme hatched by union-backed former New York City Comptroller John Liu.  He produced two reports on New York City’s public employee pensions, one that claimed that low investment returns were responsible for half the funding shortage for the city’s pension plans.  And another that claimed that the city’s high assumed future rate of return on pension assets was justified, because its past investment returns had been higher than that.  Two things that could not both be true.

PBS Frontline, in its report on public employee pension problems in Kentucky, implied by its silence that retroactive pension increases did not exist.  And by its content that Wall Street, and low investment returns, are to blame for a large share of the state and local government pension crisis.

Wall Street deserves blame for a great deal.  Including overpaying itself for the 1990s stock market bubble, the same stock market bubble that was used as an excuse to enrich state and local government pension benefits and cut pension funding, as if it was the product of the financial sector’s own genius.   And then attempting to continue to grab that inflated level of pay during the 2000s by eschewing the boring business of responsibly allocating capital and playing a giant game of poker instead, culminating in a government bailout.  With regard to the cause of the state and local government pension crisis, however, the data shows that investment returns were not the problem.

Taking the entire period from 1987 to 2019, which began and ended with stock prices at inflated levels, and dividing the state and local government pension fund investment returns reported to the Census Bureau for each year by the pension fund assets at the end of the prior year, one finds U.S. pension funds as a whole averaged a 9.7% return.   That is a very large return.

What was the average assumed rate of return during this period?   In my view it should have been no more than 7.0% on average.  Perhaps more starting at points when asset prices were low, and higher returns could be expected from those low values.  But certainly less starting at points when asset values were temporarily inflated, suppressing future returns.

Any more than that was fraudulent.   

Taking a 10-year moving average, moreover, one finds the difficult decade through FY 2019 had an average return of 10.5%.   And what about Kentucky, whose problems Frontline blamed in large part on Wall Street?  It’s average pension fund investment return for the 1987 to 2019 period was 9.1%, and its 10-year moving average stood at a more than respectable 10.0% in 2019.

Past returns were higher than should be expected going forward, not only because zero percent interest rates mean asset values are temporarily inflated, but also because inflation is low.  A large share of the past return wasn’t “real,” but was merely an artifact of inflation.   Lower inflation means lower nominal returns even if the “real” return in dollars of equal value stay the same.

Looking backward, however, for the average U.S. state and local government pension fund lower inflation, though taxing, does not explain the $3.5 trillion pension funding shortage.  As of FY 2019, the real, inflation adjusted return for the prior 10 years had been 6.9%.  With even two percent inflation, nearly a 9.0% assumed rate of return would have been achieved.  

Lower investment returns are a politically convenient scapegoat for the public employee pension crisis, but a false one.  And the assertion that public pension funds were fully funded in 2000, and that therefore there was plenty of money available to retroactively increase pensions and cut pension funding, was also false.  It was merely an artifact of a temporary asset price bubble, and the fraudulent failure to adjust future expected returns for its subsequent deflation.  The cycle has been to claim there is “plenty of money” during asset price bubbles, and then claim that “Wall Street stole our money” when asset prices deflate.  Today, however, public employee pensions remain underfunded – and taxpayer pension contributions high – despite a bubble.  No one seems interested in explaining how we ended up here. There is silence when stocks are up, and blaming returns would not be credible, followed by claims that “Wall Street” took our money, in a media massage of propaganda messaging.


Adding it up, in FY 2019 the $3.2 trillion in official state and local government debt, the $4.5 trillion in state and local government pension debt (according to the Bureau of Economic Analysis), and the $1.4 trillion in infrastructure debt, sum to a mortgage equal to 47.0% of the income of everyone on the United States, to be paid indefinitely.   That is virtually unchanged from the 47.1% of U.S. residents’ personal income I found in FY 2012, when the economy was much weaker than in FY 2019.   No progress despite an economic expansion that increased personal income, and another stock market bubble inflated by zero percent interest rates that pushed up public employee pension fund assets.   To the extent that the Sold Out Future mortgage was paid down slightly in some states and localities, it was increased in other areas.  Rising interest rates and next recession will certainly move that needle a long way in the other direction.

And we have not yet begun to pay that mortgage.  Thanks to rock bottom interest rates that both reduced the interest of state and local government debts, and inflated public employee pension paper assets, the combination of those interest payments and taxpayer pension contributions was a smaller share of personal income in FY 2019 that it had been back in the 1980s.  When a prior generation of politicians was making a prior generation of taxpayers actually pay their bills, the federal deficit caused by the Reagan tax cuts aside.  This has offset the increase in taxpayer pension funding.  Since 2009, state and local government interest plus taxpayer pension contributions have hovered just above 1.6% of total personal income.  (Increased expenditures on public retiree health insurance is on top of that).

Those who benefitted in the past are heading for senior citizen status, and retirement.  In many places that means heading for eligibility for state and local government tax breaks that will ensure they are exempted from the fallout of their own self-dealing.  In Illinois, all retirement income is exempted from state income taxes.  In New York, all public employee retirement income, and some private sector retirement income, is exempted from New York State and New York City income taxes.

The extent of the damage, the details, and the division of the blame, however, varies from state to state.  The next three posts will highlight these differences, culminating in an overall “sold out future ranking” for each state.  But one doesn’t have to wait for me to write those post to find out where their state (and my city) stand.   All the data for all the states and all the tables and charts is (once again) in the spreadsheet.  

Looking at the tabs on the bottom, it should open with the data as downloaded for every year for every state, and the calculations used in the analyses.   Several “output” tables follow, and then the charts.  The last tab is the data used in the charts.   Anyone can substitute data for their state in the first worksheet for the data for the states shown in the chart, and turn this into an analysis for their state.

This spreadsheet represents perhaps 80 hours of work, in the evening after work and on the weekends.  It is yours for the taking.

Some additional charts that will be used in the next three posts are in these spreadsheets.


1 thought on “Sold Out Futures:  A State-By-State Comparison of State and Local Government Debts, Past Infrastructure Investment, and Unfunded Pension Liabilities Through FY 2019

  1. Pingback: Sold Out Futures by State:  Debt and Infrastructure for FY 1972 to FY 2019 | Saying the Unsaid in New York

Comments are closed.