A recent Washington Post article, “As senator, Clinton promised 200,000 jobs in Upstate New York. Her efforts fell flat.”[1] points out that during Senator Clinton’s tenure between 2001 and 2009, Upstate New York saw job growth of only 0.2%, far from what Clinton claimed could be achieved. While the article neglects to point out that the nation as a whole actually lost jobs during the period, since Clinton’s term ended near the low point of the recession of 2008-2009, it is clear that her claim was unfounded.
But, Senator Clinton’s emphasis on economic development and job creation is not unique. Politicians in New York state and elsewhere regularly claim that their policies lead to job creation, often using statistics to tout their arguments. In 1994, a significant element of Governor George Pataki’s first campaign for Governor focused on his claim that the state’s loss of jobs in the period immediately prior to the campaign was a result of Governor Mario Cuomo’s tax and regulatory policies. Governor Pataki was fortunate, initially, because following the recession that took place in the early 1990’s, the national economy, and New York’s, improved. Each month for the first five years of Pataki’s terms of office, his Administration pointed to the creation of thousands of jobs in New York State.
Then, in 2000, the nation again entered recession, which was exacerbated by the 9/11 attack on the World Trade Center. Not surprisingly, New York stopped seeing job growth, and the frequent press releases ceased.
More recently we have seen Governor Andrew Cuomo point to continued job growth during his administration. In his 2016 State of the State speech, the Governor said, “We limited the state’s new spending to less than 2% a year. We passed a 2% property tax cap that has brought welcome relief to the citizens of our state and we have cut income, corporate and estate taxes. In total, we have reduced the tax burden on New Yorkers by $114 billion dollars. Why is that important? Because reducing taxes is part of our strategy to create jobs.”
During Governor Cuomo’s administration, like the first years of Governor Pataki’s administration, New York State has seen significant job growth. But can governors or senators rightfully take credit for employment growth during their administrations? Is New York’s relative job creation performance primarily the result of State and local tax and spending policy? This post will examine patterns of job growth in New York, and will attempt to provide some answers.
Employment Change in New York State and the Nation
Many analyses of employment change focus on comparisons between New York State and the national average. Between 1990 and 2015, private sector employment grew by 18.7 percent, compared with 33.5% for the nation (note, data in this report, unless otherwise noted, is from the U. S. Department of Labor, Bureau of Labor Statistics, Current Employment Statistics). When New York is broken into regions – the New York City metropolitan region, and the rest of the state (Upstate) – there is a considerable difference in performance. New York metropolitan employment grew by 24.5%, while Upstate employment grew by 6.1%.
But, a closer examination of the state’s performance shows significant variations in performance across different economic cycles. Since 1990, the nation has experienced three significant growth periods, broken by recessions in 1990-1992, 2000-2002, and 2007-2009. In the first growth period, 1992-2000, New York’s performance lagged the nation’s – private sector employment in the state as a whole grew by 15.2%, compared with 23.5% for the nation. The difference in performance between upstate New York and the New York metropolitan area was substantial – downstate employment grew by 18.2%, while upstate job growth was 8.2%.
Percent Employment Change – 1990-2015 | |||||
United States | New York | NYC Metro | Upstate | ||
Recession | 1990-1992 | -0.08% | -5.24% | -6.84% | -1.83% |
Growth | 1992-2000 | 23.52% | 15.18% | 18.70% | 8.06% |
Recession | 2000-2002 | -2.55% | -4.38% | -4.56% | -3.97% |
Growth | 2002-2007 | 6.39% | 5.10% | 6.43% | 2.14% |
Recession | 2007-2009 | -7.54% | -3.97% | -4.04% | -3.81% |
Growth | 2009-2015 | 12.88% | 12.65% | 15.54% | 6.00% |
During the second growth period – from 2002-2007, New York’s performance again lagged the nation’s, but by significantly less than in the 1990’s. Nationally, private sector employment grew by 6.4% compared with 5.1% for New York State. Employment in the New York portion of the New York metropolitan area grew by 6.4%, which was greater than the national growth, while upstate employment grew by only 2.1%.
During the third growth period, from 2009-2015, private sector job growth in New York State about equaled the growth in the nation – 12.7% in New York vs 12.9% in the nation. Growth in the New York portion of the New York Metropolitan area exceeded the nation’s – 15.54%, while that in upstate New York was again sub-par, at 6%.
New York Compared to Rust Belt States
Population growth in the United States has continued to shift south and west. That factor alone contributes to regional variations in employment change. Additionally, regions vary in “industry mix,” the relative proportions of their populations employed in different industries. Given the historic importance of manufacturing in the rust belt, states in the Northeast and Midwest have suffered more than the rest of the nation. For thirty years, manufacturing employment was stayed constant, at 18 million jobs, as service employment grew. But, the decade from 2000 to 2010 saw one in every three manufacturing jobs disappear in the United States – from 17.3 million to 11.5 million.[2]
Not surprisingly, employment growth in rust belt states in the first decade of this century reflected the weak performance of the manufacturing sector. Even before the great recession of 2007-2009 rust belt states saw little or no private sector job growth. For the rust belt, both the decade between 1990 and 2000 and that between 2010 and today saw much better economic performance.
Employment Change in Rust Belt States – 1990-2015
|
Illinois | Indiana | Massachusetts | Michigan | New York | Ohio | Pennsylvania |
1990-1992 | -0.18% | 2.45% | -3.93% | 0.81% | -5.26% | -0.34% | -1.19% |
1992-2000 | 15.78% | 17.93% | 21.10% | 20.55% | 14.56% | 17.18% | 13.50% |
2000-2007 | -1.62% | -1.04% | -2.91% | -11.85% | 0.52% | -5.22% | 1.46% |
2007-2009 | -9.09% | -10.31% | -5.11% | -12.44% | -4.48% | -9.78% | -5.55% |
2009-2015 | 9.00% | 13.41% | 11.50% | 14.45% | 12.73% | 10.91% | 7.67% |
State and Local Tax Policy and Job Creation
Does the data support the argument that state economic performance is related to tax policy? We have often seen arguments that New York, as a relatively high taxed state, is at a disadvantage to regional competitors with lower tax burdens. The data shows that some states with relatively high tax burdens – Massachusetts and New York – did better than states with significantly lower burdens – Michigan and Ohio, for example. (source -State & Local Government Finance Data Query System. http://slfdqs.taxpolicycenter.org/pages.cfm. The Urban Institute-Brookings Institution Tax Policy Center. Data from U.S. Census Bureau, Annual Survey of State and Local Government Finances, Government Finances, Volume 4, and Census of Governments (1977-2013)). It also shows that the relative performance of states varied from period to period. For example, Michigan was one of the strongest performers in the rust belt from 1992 to 2000, but was among the weakest in the recessions of 2000-2002 and 2007 to 2009.
State and Local Taxes Per Capita
Region and State | 2013 |
United States ……………………………………………………………………… | $4,599 |
Massachusetts……………………………………………………………………… | $5,723 |
New York……………………………………………………………………… | $8,047 |
Pennsylvania……………………………………………………………………… | $4,627 |
Illinois……………………………………………………………………… | $5,374 |
Indiana……………………………………………………………………… | $3,793 |
Michigan……………………………………………………………………… | $3,750 |
Ohio……………………………………………………………………… | $4,275 |
The Upstate Downstate Divide
For the past half century, Upstate New York has consistently grown more slowly than downstate, largely because of its historical dependence on manufacturing. Even so, the chart below shows that there have been significant differences in private sector employment growth between New York’s metropolitan areas. The New York City metropolitan area had the greatest employment growth – more than 25% – among those studied in New York State between 1990-2015. The Albany Schenectady Troy metropolitan area was second, with about 20% private sector job growth.
But other metropolitan areas upstate had little private sector employment growth, or in some cases, losses. Rochester’s employment grew by about 5%, and Buffalo’s by 3%. Binghamton’s employment declined by more than 15% during the period.
The job creation performance of New York metropolitan areas compared to other metropolitan areas in the rust belt varied substantially during different periods of growth and recession, even within relatively short time periods. Relative to other rust belt metropolitan areas, New York metropolitan areas showed the weakest performance in the 1990-1992 recession, and the strongest in the 2007-2009 recession. These kinds of shifts can reflect the effects of differing economic environments as they relate to metropolitan areas’ industrial bases. For example, in 2007-2009, metropolitan areas in Michigan, highly dependent on the auto industry, were particularly hard hit while New York’s metropolitan areas generally did relatively well. Syracuse and Buffalo’s performance was weak between 1990 and 2000, but did relatively well between 2000 and 2009.
Is the large variation in private sector employment change between metropolitan areas in New York State found in other states? A look at employment change in other rust belt states shows that it is.
Ohio
Michigan
Pennsylvania
The differences in employment change between cities within each state were substantially larger than those between states. For example, Columbus, Ohio metropolitan area private sector employment grew more than 40% between 1990 and 2015, while Youngstown saw a decline of nearly 10%. In Michigan, Grand Rapids private sector employment grew by more than 40%, while Flint’s dropped by nearly 20%. The high level of dispersion between the economic performance of individual cities within states points to the fact that in these historically relatively undiversified metropolitan areas, the performance of a dominant industry or company can significantly affect metropolitan area private sector employment change. Both Detroit and Flint suffered signficantly from the woes of the domestic auto industry, while the Rochester area saw Eastman Kodak employment decrease from nearly 50,000 in 1988 to a small fraction of that today.
Implications
There is clear evidence that federal policies, whether relating to labor and environmental regulations, taxes, trade, or the use of fiscal and monetary policy, can have a significant impact on corporate decision making and job growth. But, former Senator Clinton’s claims about growing the upstate economy foundered on several realities. First, the Senator failed to recognize that the region’s job creation would largely depend on national economic conditions. When the national economy contracted from 2007 to 2009, any chance that 200,000 jobs would be created in upstate New York disappeared. And, it must be recognized that as a junior senator in a body of 100 members, Senator Clinton’s influence on federal economic policy was very limited.
Policy claims about employment change in New York often center around the notion that New York’s high taxes have retarded the state’s growth. These claims are rooted in historical experience. Beginning in the 1960’s New York State began to see its manufacturing base erode, as textile manufacturers, appliance makers and others sought locations with lower labor costs and taxes, and easier regulatory policies.
But it is important to remember that even then, other factors influenced location decisions. While some people and businesses moved south and west for lower living costs, quality of life was a factor as well, probably a more important one than tax costs. People chose to locate in the sunbelt to avoid cold winters and snow, and to access new opportunities found in these areas. As the nation’s population grew in the South and West, New York and other rust belt states were no longer as competitive as locations to serve national markets as they had been. Metropolitan areas in the rust belt stagnated as areas in the South and West grew. Areas that were heavily dependent on manufacturing saw the greatest losses.
The data shows that New York’s employment change over the past 25 years has been similar to that in other rust belt states. The relatively small differences in performance at the state level do not show an association with state and local tax levels. There were large differences in relative job creation performance between metropolitan areas within states overall, and significant variations in the relative performance of metropolitan areas over relatively short time periods.
Both of these findings are inconsistent with the argument that state and local tax differences are a primary explanation of state economic performance, since state and local tax burdens within states do not significantly differ, and New York’s state and local tax burden relative to other rust belt states has not shifted significantly over time. If tax levels were a significant factor influencing job growth, we would expect to find more consistent patterns of performance within states and across time periods, and differences in job growth between states that would be consistent with differing tax burdens. Instead, the data points to the fact that job creation in metropolitan areas depends mostly on their industry mix – the performance of the companies within the industries that make up their economies.
These findings reflect the fact that today, state and local tax costs are a very small percentage of total firm operating costs, and that differences between states are even smaller. In earlier research, based on data from the Tax Foundation, I found that state and local taxes amounted to less than 4% of business operating costs (less than 2% for manufacturing businesses), on average, and that differences between New York taxes and taxes in other states were less than 2% of operating costs. These relatively small differences pale compared with the large differences in labor costs between locations in the United States and in low wage countries.
One study, by Professor Peter Navarro, estimated that differences in labor costs between the United States and China could amount to 17% of the cost of production – more than ten times the impact of state and local taxes on manufacturing operating costs. For manufacturers, the large differences in labor costs and the growth of global markets have led to the movement of manufacturing operations to locations outside the United States.
While New Yorkers might legitimately question whether the services they receive are good enough to justify paying state and local taxes that are 80% higher than the average for the nation, and substantially above the average for the rust belt, the data does not support the notion that high taxes have hurt employment levels in New York State.
________________________________________________________
Metropolitan areas included in rust belt comparison: Illinois: Champaign-Urbana, Chicago, Peoria, Rockford, Springfield; Indiana: Elkhart, Evansville, Fort Wayne, Gary, Indianapolis; Massachusetts: Boston, Springfield, Worcester; Michigan: Ann Arbor, Detroit, Flint, Grand Rapids, Kalamazoo, Lansing; New York: Albany-Schenectady-Troy, Binghamton, Buffalo-Niagara Falls, New York City, Rochester, Syracuse, Utica; Ohio: Akron, Canton, Cincinnati, Cleveland, Columbus, Dayton, Toledo, Youngstown; Pennsylvania: Allentown-Bethlehem-Easton, Erie, Harrisburg, Lancaster, Philadelphia, Pittsburgh, Reading, Scranton-Wilkes-Barre, York.
[1] “As senator, Clinton promised 200,000 jobs in Upstate New York. Her efforts fell flat.” Jerry Markon, Washington Post, August 7, 2016.
[2] Economic Policy Institute, “The Manufacturing Footprint and the Importance of U. S. Manufacturing Jobs,” Robert E. Scott. January 22, 2015. http://www.epi.org/publication/the-manufacturing-footprint-and-the-importance-of-u-s-manufacturing-jobs/
This is a very fine analysis of some of the structural and industry mix differences that have played a role in the economic performances of the Rust Belt states, including New York. Unfortunately the author has, in my view, muddled it a bit by concluding that his findings somehow refute the idea that tax levels are “a significant factor influencing job growth.”
The experiences of the Rust Belt states vis-a-vis each other cannot be expected to tell us much about the tax question, one way or the other. No company seeking lower costs, or worried about competitors in other states who have lower costs, is thinking about the Rust Belt. To be frank, nobody moves from New York to Illinois, or from Ohio to New York, in pursuit of lower taxes and a better business climate – they look outside the region entirely. And when they do, they will find bigger differences than the author credits here.
In the most recent 20 years for which apples-to-apples data on jobs and taxes are readily available (1992-2002), Texas grew jobs more than four times as fast as New York. Florida and Colorado grew three times as fast, and Georgia, North Carolina and Virginia all grew at better than twice New York’s pace. All of these states have state and local tax burdens below the national average. The ten states (other than New York) that added a half million or more jobs each in that period had state and local tax burdens about four percent below average.
What does this prove? It does not prove that taxes entirely explain the difference, any more than the experience of Rust Belt states proves that taxes are NOT a factor. But as far as I know, neither Governor Cuomo nor any other responsible party has ever contended that tax levels are the sole or even controlling factor in states’ economic performances. The weather is certainly a factor, as is proximity to markets – as the author notes. So, too, are the cost and availability of the needed workforce, and the regulatory environment. (I suspect these latter two, by the way, are to some extent linked to higher taxes, in a curiously attitudinal way – a political culture that thinks government is so good that it is entitled to an inordinate share of our income will also tend to think that it should do things like mandate higher wages and impose an expansive regulatory regime.)
Finally, the author should be commended for noting that a central issue with respect to the tax burden is not the (disputed) economic impact, at all. It is, instead, the question of whether taxpayers are getting what they are paying for. If you are of a humanitarian outlook and think, for example, that higher taxes in New York should enable our state and local governments to provide top-drawer services to those who most need our help, look no further than the annual Kids Count databook produced by the Annie E. Casey Foundation; it ranks New York 25th among the states in terms of child well-being. If high taxes bought superior infrastructure, our international airports wouldn’t have that third-world look to them, and our key transportation corridors wouldn’t be toll roads. If high taxes produce better schools, we should have seen the results billions upon billions of dollars ago.
To me, this means that the last thing New York’s taxers need is encouragement.
David – Thank you for sharing your perspective. I’d like to respond to a few of the points you made in your response. Your primary criticism of my analysis had to do with my choice of states – those in the rust belt – with which I compared New York’s performance. While you agree with the argument in my paper that other factors such as labor costs, climate, and locations nearer to markets, you contend that lower taxes are important.
For your argument, you included states to the west and south, Texas, Colorado, Florida, North Carolina and Virginia which are growing significantly more quickly than New York, and argued that “all of these have state and local tax burdens below the national average.”
I would agree that these states are growing faster than New York, but I question the causality argument that is implied in your statement. I’m troubled by your choice of a few low tax states that happen to fit your argument while excluding others. While some low tax states have grown more quickly than New York, a recent Pew Report (http://www.pewtrusts.org/en/research-and-analysis/blogs/stateline/2015/5/13/which-states-have-the-most-job-growth-since-the-recession) shows that among the slowest growers have been a number of low tax states – Mississippi, West Virginia, Alabama, Arkansas, Missouri, are examples.
I would argue that the fact that some low tax states have been growing more slowly than the nation, and that others have been growing more quickly points to the fact that factors other than low levels of state and local taxes are the primary drivers of employment growth. And, as I noted in my piece, the relative performance of states varies significantly depending on the beginning and end points chosen for an analysis.