The release of Bureau of Economic Analysis Local Area Personal Income Data for 2019, following the earlier release Current Employment Survey data from the Bureau of Labor Statistics, and American Community Survey data from the U.S. Census Bureau, completes the chronicle of the period from 2007 to 2019, from the peak of one economic boom to another.
And what a boom it was for New York City. Not since the 1920s has NYC prospered as much compared with the nation as a whole, and back then large areas of the city were still undeveloped, allowing it to expand geographically. Kings County (Brooklyn) added the most new housing of any U.S. county that decade. By 2007, in contrast, the NYC metro area was either densely developed, or developed and restrictively zoned, 75 miles out from the center in all directions. So for the first time since the start of the suburbanization era growth was concentrated in the center, in NYC and urban areas such as Hudson County, New Jersey. As we eventually face the recession and fiscal crisis that was due before the pandemic even hit, it is worth remembering that Governor Cuomo, Mayor DeBlasio, the NY state legislature and NYC council benefitted from a gusher of tax revenue that should have provided a foundation for a better future. If it didn’t, we were robbed.
An updated spreadsheet with the charts used in this post is here.
The New York Metro Area may be defined in different ways. For purposes of this data I’m leaving New Haven County in Connecticut and Dutchess County and Orange County in New York out, with the latter in “Upstate Urban Counties” and the former in “Rest of Connecticut.”
Looking at data for the peak year of the last few economic expansions, metro New York continues to account for a shrinking share of the U.S. population. From 7.2% in 1979, Metro New York’s share fell to 7.3% in 1987, 6.8% in 2000, 6.5% in 2007, and just 6.2% in 2019. In straight numbers, the population rose from 17.3 million to 17.8 million, 19.3 million, 19.5 million, and 20.2 million. As the soaring housing prices of the past 20 years, despite an urban development boom, shows, it would be difficult for Metro New York’s population to grow faster than that.
From 2007 to 2019, the population of the United States increased by 9.1%. The percentage gain was 4.0% for New York City, 2.3% for the Downstate Suburbs, 4.3% for Northern New Jersey, 3.1% for Central New Jersey, and 5.1% for Fairfield County. Meanwhile the total population of the urban counties Upstate increased by just 0.2%, while the population of rural Upstate fell 3.9%. The population of the Rest of Connecticut, excluding Fairfield County, fell 0.3%.
Metro New York’s share of U.S. work earnings by place of work is much higher, however, and it has decreased much more slowly. After rising from 9.1% of the U.S. total in 1979 to 10.1% in 1987, it slipped to 9.7% in 2000, 9.4% in 2007, and 9.3% in 2019. Metro New York thus accounted for a larger share of total U.S. earnings at work in 2019 than it had in 1979, forty years earlier. That shows just how hard Metro New York was hit in the 1970s. (Metro NY had accounted for 11.1% of U.S. work earnings in 1969).
Remarkably, New York City also accounted for a higher share of U.S. earnings by place of work in 2019 than in 1979. NYC accounted for 4.3% in 1979, 4.7% in 1987, 4.6% in 2000, 4.5% in 2007 and 4.8% in 2019. A greater share of the nation’s work earnings has thus been concentrated in a city whose boundaries have not expanded since 1898. (NYC had accounted for 6.0% of total U.S. earnings by place of work in 1969, and presumably more before that). In 2019 NYC accounted for 51.5% of the work earnings by place of work in Metro New York.
New York City, on the other hand, only accounted for 38.7% of Metro New York work earnings by place of residence, since a substantial share of the money earned at work in NYC is paid to those who commute in. Some of that money may now be earned at home, or at satellite offices in the suburbs, in the wake of the pandemic. It is already exempt for NYC local income taxes as a result of the repeal of the commuter tax, but there may be fewer commuters from New Jersey and Connecticut paying New York State income taxes in the future.
Otherwise the pattern is the same as for work earnings by place of work. Metro New York has accounted for a shrinking share of the U.S. total since 1987, but still accounts for a higher share than in 1979. New York City residents’ percent of total U.S. work earnings was 3.2% in 1979, 3.6% in 1987, 3.4% in 2000, 3.2% in 2007 – and 3.5% in 2019. Clearly the 2007 to 2019 period was a strong one for the work earnings of city residents, as a rising share of high paid U.S. jobs concentrated in New York City, and a rising share of those who held such jobs lived within city limits. As noted in a prior post, American Community Survey data shows that from 2000 to 2019 the number of employed residents of NYC increased by about 850,000.
Work earnings is one of three components of personal income, as recorded by the Bureau of Economic Analysis. The others are actual cash income from investments, dividends, interest and rent (as opposed to paper gains from rising asset prices). And “transfer payments” from the government to the household sector, mostly for the old (Social Security, Medicare, Medicaid) but also for the poor (SSI, SNAP, Medicaid also, unemployment insurance payments) etc.
In 2019, work earnings accounted for 62.9% of U.S. personal income, compared with 61.1% for residents of New York City, 63.8% for the Downstate Suburbs, 69.0% for Northern New Jersey, 68.1% for Central New Jersey, 62.6% for Fairfield County, and 66.2% for the Rest of Connecticut. Work earnings accounted for 62.4% of personal income in urban Upstate New York, but just 59.3% of personal income in rural Upstate New York. I would expect work earnings to account for a relatively low share of personal income in rural areas of the U.S. in general, due to aging populations.
Dividends, interest and rent accounted for 20.2% of U.S. personal income. New York City was above average with a 22.9% share, as were the Downstate Suburbs at 23.8% and Fairfield County at 29.9%. Northern New Jersey at 18.7%, Central New Jersey at 18.3%, the Rest of Connecticut at 16.6%, Upstate Urban Counties at 17.5%, and Upstate Rural Counties at 16.3%, all had a relatively low share of their residents’ personal income come from investments.
Transfer payments accounted for 16.9% of U.S. personal income in 2019. Residents of the suburbs around NYC had a below average share of their personal income come from this source, at just 12.5% in the Downstate Suburbs, 12.2% in Northern New Jersey, 13.6% for Central New Jersey, and 7.5% for Fairfield County. Since a large share of transfer payments goes to the old, not the poor, this is in part a consequence of the large number of New York area seniors who move elsewhere in retirement. Transfer payments accounted for 19.2% of total personal income for residents of the state of Florida in 2019.
New York City has a lingering reputation as a “welfare city” with lots of poor people, and lots of government spending on the poor. It still has an above average percent of its population living in poverty, but transfer payments accounted for just 15.9% of its residents’ total personal income in 2019. The white seniors who were left behind as the middle class suburbanized from the 1950s through the 1970s have died off, reducing spending on old age benefits, and from 2000 to 2019 a large number of young working “Millennials” moved in, increasing work earnings. Some of them came from Upstate New York, where a stagnant and aging population has increased the share of personal income accounted for by transfer payments to 20.1% in urban counties and 24.5% in rural counties, far above the U.S. average.
During the most recent economic cycle, from 2007 to 2019, total inflation-adjusted work earnings by place of residence increased by a stunning 31.3% in New York City, well above the 19.6% increase for the U.S. as a whole. Retiring Baby Boomers limited the increases to 12.8% in the Downstate Suburbs, 10.9% in Upstate Urban Counties, 6.6% in Upstate Rural Counties, 17.7% in Northern New Jersey, 12.8% in Central New Jersey, 7.6% in Fairfield County, and just 0.8% in the Rest of Connecticut.
On the other hand, the total inflation-adjusted transfer payments received by NYC residents increased by just 22.1%, far less than the 46.7% increase for the U.S. as a whole. The U.S. increase was driven by a rising share of the population exceeding 62-65 years old and beginning to collect Social Security and Medicare benefits, and a rising share of the workforce falling out of the middle class and joining the working poor, eligible for benefits such as SNAP (Food Stamps), the Earned Income Tax Credit, Medicaid, and Obamacare.
Around Metro New York and Upstate, transfer payments increased more rapidly than in New York City, but more slowly than the national average. The increase was 37.3% for the Downstate Suburbs, 36.5% for both the Upstate Urban counties and the Upstate Rural counties, 30.9% for Northern New Jersey, 37.7% for Central New Jersey, 35.1% for Fairfield County and 37.3% for the Rest of Connecticut. Not every retiree, it would seem, has moved to Florida.
I was surprised to find that inflation-adjusted investment income increased 28.6% in the U.S. from 2007 to 2019, and 39.6% in New York City. After all interest rates have been very low during this period, and so have dividend payments by companies, as a percent of stock prices. There were also gains of 33.0% for Upstate Urban Counties and 37.4% for Upstate Rural counties. Could this represent increased rental income, as more people rented out spare rooms and housing units on platforms such as AirBNB?
Perhaps, but in that case I would have expected a greater increase in Rural New York compared with Urban New York.
Another possibility is that the dividends, interest and rent of seniors with tax-deferred retirement accounts are not counted until the money is withdrawn and taxed. In that case, the very affluent generation of seniors now in retirement and living off their accumulated savings would tend to push recorded investment income up.
One issue in Upstate New York is the dependence of the consumer economy on spending by the well off seniors who once held the high paid jobs that used to be located there. Those jobs have been disappearing for decades, and the related retirement income will eventually disappear as well. So the region will become poorer and suffer from diminished consumer demand when today’s retirees die off.
For those who want to examine these issues in more detail, the BEA does provide more local detail for transfer payments, but not for dividends, interest and rent.
Despite the varying trends for transfer payments and investment earnings, Metro New York’s share of total U.S. personal income by place of residence has followed the same pattern as work earnings. That share was 9.1% in 1979, 9.6% in 1987, 9.4% in 2000, 9.1% in 2007, and 8.9% in 2019.
Amazingly, New York City residents’ share of total U.S. personal income was 3.6% in 2019, up from 3.2% in 1972. New York City had accounted for 4.9% of total U.S. personal income in 1969, another indication of what the decade of the 1970s was like for the city, as the middle class fled to the suburbs leaving behind the poor, the old, city debts, a deteriorated infrastructure, and the need to pay for retroactively increased pension benefits. Overall, NYC residents’ share of total U.S. personal income at the peak year of the economic cycle has been relatively stable since 1987.
Taken together, the suburbs around New York City accounted for 5.6% of total U.S. personal income (by place of residence) in 1979 and 6.0% in 1987, followed by 5.9% in 2000, 5.7% in 2007 – and just 5.3% in 2019. While the 2007 to 2019 period was booming for the city, it was far less prosperous for the suburbs.
Although work earnings per worker have been weak for decades, particularly outside the top one percent, total inflation-adjusted U.S. personal income per capita has continued to rise. The share of adults who were working increased until the Great Recession, when many Baby Boomers started to move into retirement. Since then the richest generations in U.S. history have moved past age 62, replacing the poorer generations that preceded them, and increasing investment and transfer payment income. With more working adults, and then more retired adults, the share of the population that is children has been going down. Basically, per capita income has increased in large part because there are more income earning (or receiving) adults, and fewer non-earning little capitas.
Long term this is not good news. In the past while many developing countries suffered from excessive population growth, and many developed countries were faced with birth rates far below replacement, the U.S. had a birth rate right around replacement. Today, however, both the U.S. birth rate and immigration have plunged, a judgment on the ability of today’s workers to support families in the new economy, and on their hopes for the future. One wonders who will pay for the old age transfer payments of today’s workers when that future arrives.
The per capita income of New York City residents was 25.9% higher than the U.S. average in 1969, but just 10.1% above average in 1980. The 1980s recovery brought it back to 26.0% above average in 1989, and it was mostly stable at 20.0% to 30.0% above average from 1990 to 2007. In 2019, however, NYC’s per capita income was 42.2% higher than the U.S. average. That gap might possibly be the largest on record, given than even back in the heydays of the 1920s NYC was a center of low wage manufacturing – think the garment industry – and low wage recent immigrants.
The per capita income of residents of the Downstate Suburbs has remained high, even as the work earnings of those living and working there has slipped. The high investment earnings of affluent retirees took their place, and with many of their children now adults and living and working in NYC in 2019, there were fewer non-earning children to bring per the suburban capita income down. In 2019, the per capita income of the Downstate Suburbs was 55.0% higher than the U.S. average. That is a record gap according to this data set, which starts in 1969. The per capita income of Northern New Jersey residents was 34.2% higher than the U.S. average in 2019, slightly lower than the 37.8% above average in the year 2000.
Upstate New York has not fared as well. The per capita income of residents of the urban counties there was 5.4% above the U.S. average in 1989, the peak year, but 2.8% below average in 2019, reflecting a downward trend for former manufacturing powerhouses that is even more pronounced in the “Rustbelt” states of the Midwest. The per capita income of Michigan was 4.5% above the U.S. average in 1969 and 3.6% above average in 1979, but it was 12.9% below average in 2019. Ohio was 0.7% above average in 1969 and 1.6% below average in 1979, but 11.1% below average in 2019. These states don’t have a New York City to suck state tax revenues out of.
In the Upstate rural counties per capita income was 13.3% below average in 1969, but 21.1% below average in 1978, worse than the 18.3% below average in 2019. Rural New York, rural America in general, looks better compared with the U.S. average when the economy is down, because transfer payments continue to be paid even as work earnings shrink. Nonetheless, the Upstate Rural economy was much stronger, relative to the country, before 1970.
The per capita income of the residents of the Rest of Connecticut was 24.9% above the U.S. average in 1989, just after Mystic Pizza was filmed there. The 1980s were very good to that state. It was just 8.7% above average in 2019.
New York City’s above average per capita income is largely a function of the very large number of very wealthy people residing in Manhattan. In 2019 the only other borough whose per capita income was above the U.S. average was Staten Island, at just 4.2% above average – down from 20.1% above average in 1990. Queens has taken an even greater fall, from 30.0% above the U.S. average in 1969, when it was a recently-built semi-suburban borough, to 6.2% below average in 2019, as a home for rising first generation immigrants.
The big mover in recent years, however, has been Brooklyn. It was never a wealthy place, but in 1969, before the decline of the 1970s, its residents’ per capita income was just 1.6% below the U.S. average. That gap didn’t peak until 2001, when Brooklyn’s per capita income was 19.6% below the U.S. average. Even in 2007, the peak of the prior economic expansion, it was still 16.7% below average. By 2019, however, Brooklyn’s per capita income was just 0.7% below the U.S. average, all the way back to where it had been in 1969.
The per capita income of Bronx residents, meanwhile, had been just 7.3% below the U.S. average in 1969, but it was 34.8% below average in 2000. The improving fortunes of NYC in the two decades since had only modest impact here. It was still 29.7% below average in 2019.
The per capita income of New York County, aka Manhattan, was once again the second highest in the U.S. in 2019, at $197,847 per person. With many lower paid young workers and students having left during the pandemic, it will no doubt be even higher next year. That per capita income is so high in a place with so many people is amazing. Many of the counties near to the top are resort areas for the rich such as Teton (Jackson Hole, in the top spot), Pitkin (Aspen), and Nantucket. These areas have small populations, so a few rich residents can really pull per capita income up. San Francisco, the western equivalent of Manhattan, was sixth, with many suburban counties in the New York metro area and Bay Area among those in the top 20. These include Fairfield County at number 11, with a per capita income of $121,397 in 2019.
Manhattan residents had a per capita income that was around 3.5 times the national average each year from 2017 to 2019. From 1999 to 2011, Manhattan had been in the vicinity of 3.0 times the national average or less, whereas from 1969 through most of the 1980s it had been merely double the national average, or less. Always a center of wealth, Manhattan had become even more so in recent years.
Because per capita income is an arithmetic mean, it is highly influenced by the massive incomes of those at the very top. Median household income, which measures the situation of the middle persons living in each place, is highest in suburban Washington DC, not in Manhattan.
At one time, in fact, Manhattan was not that rich at all. While there were plenty of rich people in Manhattan in 1950, for example, the Census of Population for that year showed they were not representative or the borough as a whole. The median household income recorded for the year before was $2,347 in Manhattan (obviously not inflation-adjusted to the present), less than the $2,619 for the United States, and well less than the $3,151 for Brooklyn, the $3,297 for the Bronx, the $3,817 for Queens, and the $3,443 for Staten Island. This despite the per capita income of Manhattan being higher than the other boroughs, even back then, thanks to a few Rockefellers, Whitneys and Astors, etc.
This shows how New York could still be a well off and dynamic city, even while being a different city, in the wake of the pandemic. Somebody will fill those Manhattan apartments if the rent falls low enough. Probably those who work there in jobs that are not among the highest paid, as in the pre-suburbanization era.
Finally, in a prior post with state-level data from the Bureau of Economic Analysis…
I showed that most of the states that are most-reliant on government spending – either transfer payments or work earnings from government employers – were Republican voting so-called “Red States.” And it is these states that have fared better in 2020, because those transfer payments have continued (and in fact have increased), while work earnings have plunged. Meanwhile, the states with a high share of personal income accounted for by private sector work earnings, such as Massachusetts and California, have been hit hard. The same has been true within the Tri-State area.
The data shows that nationwide, work earnings in state and local government accounted for 11.9% of total work earnings in 2019. Government accounts for a higher share of the total in recessions, as evidenced by the 13.9% in the year 2009 during the Great Recession, and less during booms. State and local government had accounted for more than 11.9% of total work earnings each year since 1989; it had generally been lower previously.
State and local government accounted for just 10.8% of total earnings at work in New York City in 2019, less than the U.S. average. A large share of those earnings, moreover, went to residents of the suburbs, not the city, since it is the highest paid titles (police, fire, correction, teacher) that tend to live elsewhere and commute in. An extremely high share total of NYC local government earnings, moreover, is in the form of retirement benefits, for those now retired to Florida or the Carolinas. Even so looking back, there are only a few years for which state and local government earnings were a (slightly) higher share of total work earnings at work in NYC than the national average.
In the other parts of New York State, the Downstate Suburbs, the Upstate Urban Counties, and the Upstate Rural Counties, on the other hand, state and local government earnings have consistently been a higher share of earnings at work than the U.S. average. In the Upstate Rural Counties, as I have shown, there are just 6.5 public school students per public school instructional employee.
With some ups and downs, however, state and local government earnings have been stable as a share of the total since 1992 in the part of New York State outside New York City, even as the earnings per worker for state and local government has soared compared with the private sector. This implies that beyond a certain point, as state and local government workers take more and more compared with private sector workers, the damage shifts from tax increases to service cuts.
In 2019 state and local government earnings were about average as a percent of total earnings at work in Central New Jersey, and below average in Northern New Jersey and Fairfield County. They were above average in the Rest of Connecticut.
New York City has long been associated with public welfare, and looking at transfer payments as a percent of total resident personal income, one can see why. NYC had the highest percent of any area charted until the late 1990s, when it was surpassed by the Upstate Rural Counties. Starting in the Great Recession, the Upstate Urban Counties also became more dependent on transfer payments than NYC. In these areas, an increase in the share of seniors as a percent of the population means an increase in Social Security, Medicare and custodial care funded by Medicaid as a percent of personal income. During the Great Recession, many former Upstate workers also ended up on SSI disability.
NYC fell below the U.S. average – and the average for the portion of Connecticut outside Fairfield County – in 2018 and 2019. The suburbs around NYC remain less dependent on transfer payments than the city itself, but (with the exception of Fairfield County) are more dependent than they once were.
The high level of state and local government earnings in Upstate New York and the Downstate Suburbs, the high level of transfer payments Upstate, the extensive state and local government earnings within NYC that is paid to residents of the suburbs, whose politicians thus seek to inflate to the extent possible, and city taxes diverted to private sector contractors and unions in the city and suburbs, in part explains NYC’s fiscal crisis despite this incredible boom. Don’t blame the poor, or even later-hired city workers now living within city boundaries.
It is only the extensive work and investment income of those residing in Manhattan, however, that has recently pushed transfer payments to a below average percent of personal income in NYC. And only Brooklyn has shown a modest downward trend, adjusted for the business cycle. The nation’s most expensive Medicaid program drives total transfer payments in the outer boroughs of New York City, and has since the days when “Medicaid Mills” popped up within a few years of the program’s creation.
So now what?
Sooner or later, the coronavirus pandemic will end, and New York City will recover some of the economic activity it has lost. But the extreme concentration of work earnings in the city relative to the rest of the metro area, and in a few metro areas (New York included) relative to the rest of the country, was never going to be sustainable. The pandemic has merely accelerated a shift, to the detriment of NYC, that was bound to take place eventually anyway. A return to the disaster of the 1970s, or even the early 1990s, however, can only occur as a result of public policy at the federal, state and local level.
If they are to avoid killing the golden goose and losing the cash cow, the federal government, other parts of New York State, and New York City’s public employees are going to have to suck far less out of New York City residents and businesses than in the past. They have to give the serfs a break, but are unwilling to do so, and in fact expect to suck out even more, and more, and more, because that is what they believe they deserve. And that is what they will continue to believe, and what will continue to happen, until someone is willing to pay the political (or cancel my subscription) price of telling them a truth they do not want to hear. Someone other than a blogger with no subscriptions, no advertisers, no revenues, no public or government-related job, and no need for votes.