For a print copy of the technical and data documentation for the NYC-STAMP model, send written request via email to the Beacon Hill Institute (firstname.lastname@example.org).
In the last five years, New York City’s economy has boomed and private sector employment has hit record levels. What, if anything, did lower taxes have to do with these achievements? And what are the implications for future tax policy?
Using an econometric model, NYC-STAMP, we reach these conclusions:
- Reductions in the City’s personal income, sales, business and property taxes have generated more than 80,000 new jobs since 1997, or about one of every four gained by the City during that period;
- More than 6,500 new jobs will be generated by tax cuts that were included in the City’s fiscal 2002 budget but are still awaiting the state Legislature’s approval;
- Nearly 15,000 more jobs could be added to New York’s employment base by eliminating what’s left of the personal income tax surcharge first adopted by the City a decade ago;
- Thanks to these additional jobs, New York City’s job growth rate exceeded the national average—the first time that has happened during an economic expansion since 1950.
The NYC-STAMP model also can be used to predict the consequences of reversing tax cuts.
- Undoing the recently enacted cut in the income tax surcharge would reduce employment growth by over 6,300 jobs;
- Full restoration of the former 12.5 percent personal income tax surcharge would result in the destruction of nearly 25,000 jobs;
- Restoring both of the Dinkins-era surcharges would cost the City nearly 37,000 jobs.
The lesson for the City is clear: tax cuts create jobs, tax increases kill jobs.
This study presents the findings of a tax model commissioned by the Manhattan Institute and adapted by the Beacon Hill Institute at Suffolk University in Boston, Mass., from its State Tax Analysis Modeling Program (STAMP). The author gratefully acknowledges the assistance of Professor David G. Tuerck, Executive Director of BHI and Chairman of the Economics Department at Suffolk University, in preparing this report.
WHAT NEW YORK HAS GAINED FROM TAX CUTS
Results of a New Econometric Model
New York City has taken some historic steps to reduce its notoriously heavy local tax burden in recent years. As City tax cuts were phased in, New York’s economy boomed and private sector employment hit record levels.
What, if anything, did lower taxes have to do with the City’s economic gains? And what are the implications for future tax policy?
To answer these important questions, the Manhattan Institute has developed a New York City variant of a proven econometric tool—the State Tax Analysis Modeling Program (STAMP).
Our model, known as NYC-STAMP, reaches these conclusions:
- Reductions in the City’s personal income, sales, business and property taxes have generated more than 80,000 new jobs since 1997, or about one of every four gained by the City during that period.
- More than 6,500 new jobs will be generated by tax cuts that were included in the City’s fiscal 2002 budget but are still awaiting the state Legislature’s approval.
- And nearly 15,000 more jobs could be added to New York’s employment base by eliminating what’s left of the personal income tax surcharge first adopted by the City a decade ago.
The NYC-STAMP model also can be used to predict the consequences of reversing tax cuts. For example, the model estimates that full restoration of the former 12.5 percent personal income tax surcharge would result in the destruction of nearly 25,000 jobs. Restoring both of the Dinkins-era surcharges would cost the City nearly 37,000 jobs.
Our model shows that tax cuts are the reason why New York City has added private sector jobs faster than the national average over the past three years. If the tax cuts we analyzed had not been enacted, job growth in the City would still have been quite strong by local standards—but, as in the past, below the national average.
The lesson is clear: Tax cuts work. And in years ahead, the best way to continue building New York’s job base will be to continue reducing its still-heavy tax burden—and, by all means, avoid adding to it.
History’s Lesson: Tax and Spend=Boom and Bust
New York City taxes have been reduced by a total of $3.1 billion over the past seven years, including more than a half-billion dollars worth of property and income tax cuts fully subsidized by the state’s School Tax Reduction (STAR) program. A new tax cut package with a total budgeted impact of $500 million this year was adopted by Mayor Giuliani and the City Council as part of the fiscal 2002 City budget, although part of the package still needs state legislative approval.
These huge tax reductions represent a striking turnabout. For most of the post–World War II era, New York City taxes headed in one direction—up—with the biggest increases taking place in the 1960s and early ‘70s. Yet these tax hikes not only failed to prevent the City from going broke—they arguably contributed to the massive loss of jobs and businesses that brought the fiscal crisis to a head in 1975.
In the aftermath of the 1970s crisis, the Koch Administration enacted a series of relatively modest, targeted reductions in business taxes. Responding to federal tax changes, the City also launched a reform of its own income tax structure. But when fiscal push came to shove with the economic slowdown of 1990, the City starting raising taxes again—enacting, in quick succession, two surcharges that added more than 28 percent to personal income tax bills, and a major property tax hike. A 1991 study by then-City Comptroller Elizabeth Holtzman predicted that over 100,000 jobs would be lost as a result of these increases. In fact, employment dropped by over 300,000 before the economy hit bottom in 1993.
A review of New York City tax policy and economic conditions over the past three decades suggests this was no accident. A clear pattern emerges: tax increases coincide with job losses, and tax cuts coincide with job gains. This is especially true when the value of state income tax cuts is factored into the mix. For example, as illustrated below in Figure 1, private sector job growth in the City has generally surged following cuts in the combined state and City income tax rate, which now stands at its lowest point in 35 years.
The link between taxes and jobs was further highlighted in a recent economic study of revenue trends in four major cities—New York, Philadelphia, Houston, Minneapolis. The study found, among other things, that a large share of New York’s job losses since 1970 could be attributed to increases in its income tax during that period. Conversely, the same model suggested that the 1998-99 cuts in New York’s top personal income tax rate would result in a gain of over 50,000 jobs.
One of the study’s co-authors, Professor Robert Inman of the University of Pennsylvania’s Wharton School, recently estimated that the combined effect of all City tax cuts over the past few years has been to boost private employment by 4 percent—about 100,000 jobs.
While the findings of Inman and his colleagues are helpful in building an understanding of the role of tax cuts in promoting growth, their study is focused primarily on defining the point at which cities approach their practical tax limit. A different, more versatile tool is needed to measure and predict the impact of City tax policy on an ongoing basis. That tool is NYC-STAMP.
The NYC-STAMP Model
The general purpose of an econometric model is to describe how an economy is affected by changes in factors such as taxes, regulations, interest rates and labor costs. This study features a model designed to explain how New York City’s economy responds to increases or decreases in City taxes. Standard statistical methods are used to estimate the significance of relationships between economic variables (such as employment) and tax variables (such as the marginal income tax rate). The variables themselves are drawn from a quarter-century’s worth of City economic and tax data (see Appendix for more details).
The State Tax Analysis Modeling Program—STAMP—was initially created by the Beacon Hill Institute at Suffolk University in Boston to measure the effects of income tax cuts in Massachusetts. It has been updated almost annually for the Commonwealth, and similar models have been adapted to almost a dozen other states.
The Manhattan Institute commissioned and participated in the development of a New York City version of the model, NYC-STAMP. The model estimates the impact on employment of changes in four categories—income tax, sales tax, property tax and general corporation tax.
The NYC-STAMP model has the added feature of re-estimating revenue impacts of tax changes on a “dynamic” basis, which takes account of both the immediate revenue loss and the revenue gained from new employment and economic activity generated by the tax cuts. Thus the model can quantify two crucial and often hotly debated issues—the extent to which tax cuts “pay for themselves,” and the extent to which tax increases fail to raise as much money as expected.
Our Findings: Tax Cuts Create Jobs
Our analysis focused on income, sales, property and business tax cuts phased in by the City starting in 1997.
The model finds that tax changes produced 80,444 private sector jobs. According to the model’s estimates, that total consists of:
- 46,262 jobs from personal income tax cuts, which have included expiration of the 12.5 percent surcharge, rate cuts and credits subsidized by the state through the STAR program, a new resident unincorporated business tax credit, and the partial rollback of the 14 percent surcharge;
- 13,792 jobs from cuts in the sales tax, including an exemption for all clothing and footwear purchases under $110;
- 10,463 jobs from property tax reductions, including state-subsidized STAR exemptions and co-op and condominium abatements; and
- 9,926 jobs from business tax cuts.
Of course, these tax cuts were enacted and phased in at the height of a national economic expansion, a time when New York City’s own growth prospects had been further bolstered by improvements in its quality of life—especially by a huge drop in crime. In addition, as the City emerged from the severe recession of the early 1990s, office vacancy rates were high and rents were relatively low, at least by New York standards. In short, skeptics would argue, the City was poised to enjoy an employment boom regardless of what changes it made in tax law.
In fact, our analysis does not find that tax cuts were responsible for all of the job gains, or even most of them. However, we find that tax cuts were the reason why New York City has managed to add private sector jobs faster than the national average since 1997. In fact, the last four years have produced the strongest job growth on record in New York City.
Our model indicates that if the City’s broad-based tax cuts had not been enacted, its private sector employment base would have expanded by 8.2 percent over the past four years—an exceptionally strong performance, by historic standards. But jobs attributable to tax cuts pushed the City’s private job growth to 10.95 percent, compared to a national growth rate of 8.4 percent.
This is truly a precedent-shattering development. In every prior economic expansion since 1950—including a couple of roaring bull markets on Wall Street—New York City trailed annual job growth in the rest of the country. The difference, this time, has been tax cuts.
The Real Revenue Impacts of Tax Cuts
Some tax cut critics acknowledge the beneficial economic effect of reducing taxes, but complain that the City has still lost too much revenue in the process. They would point to official estimates that the broad-based tax cuts cited above are costing the City $2 billion a year in foregone revenue. However, these estimates overestimate the true revenue impact because they are “static,” in that they assume tax cuts have had no impact on taxpayer behavior. Since economists of all stripes agree that tax cuts generally stimulate economic activity, the true amount of foregone revenue must be lower than official estimates.
The NYC-STAMP model is able to re-estimate the revenue impacts of tax changes on a “dynamic” basis, taking account of both the immediate revenue loss and the revenue gained from new employment and economic activity generated by the tax cuts. According to NYC-STAMP, these tax cuts will have a net cost of $1.6 billion in 2001—or about $400 million less than has been assumed.
Benefits of New and Pending Tax Cuts
Keeping and adding jobs is increasingly important as the national and City economies slow. Thus, we have also chosen to model the employment effects of three recently enacted tax cuts whose continued survival may be in doubt, depending on the policies of the next mayor and the inclinations of the state Legislature.
The adopted fiscal 2002 budget includes a further reduction in the personal income tax surcharge, which took effect July 1. However, the City is still awaiting state legislative approval for two additional tax changes included in this year’s budget: elimination of the remaining sales tax on clothing, and an added tax abatement for co-ops and condominiums. If approved, NYC-STAMP finds, these changes will increase employment by an additional 6,643 jobs.
Looking Ahead: Tax Hikes Will Kill Jobs
As the economy slows and the accumulated City budget surplus is depleted, many fiscal monitors predict multi-billion dollar budget gaps for each of the next four years. Mayor Giuliani strongly disputes the most pessimistic forecasts. The City Council staff also has predicted higher revenue estimates and smaller gaps over the next four years. Nonetheless, even under the most optimistic scenarios, it is clear that the City budget is about to become tighter than it has been at any time since 1994–96. If history is any guide, pressure may grow to derail planned tax cuts or even return to a path of increasing City taxes to pay for new or additional spending.
In light of this possibility, we used the model to assess the impact of three potential tax increases—reversing this year’s income tax cut, reviving the 12.5 percent personal income tax surcharge, and restoring both income tax surcharges in their entirety.
Undoing this year’s income tax cut—which shaved away another portion of the remaining surcharge—would reduce private employment growth by 6,335 jobs, our model indicates.
Re-applying the 12.5 percent surcharge to current tax rates would have the effect of reducing employment by 24,631 jobs, according to the model. The dynamic revenue gain would be $612 million—$68 million (or 10 percent) less than would be assumed using standard revenue estimation techniques.
Completely restoring both surcharges on the new base would destroy 36,747 jobs and would raise $104 million less than a typical static revenue estimate would indicate.
Conversely, what benefits might the City expect from continuing to cut taxes? One worthy objective—elimination of the surviving portion of the 1991 surcharge—would generate an additional 14,927 jobs, as further explained in the Appendix to this report.
As a world financial capital, New York City is greatly affected by global and national economic conditions. But New York City itself is an open economy—which means it is also, to a great extent, the master of its own economic fate.
Businesses and households can and will depart for greener pastures if the perceived costs of staying in New York outweigh the perceived costs of relocating someplace else. Indeed, many have found that they can draw on the comparative advantages of working and doing business in New York while minimizing their physical presence within its borders.
No one would seriously argue that taxes are the only factor influencing the City’s economic condition. However, our model shows that tax cuts deserve credit for about one-fourth of New York’s recent job growth—not an insignificant share, by any means.
In other words, at the height of a national economic expansion, an aggressive tax-cutting policy made the difference between very good employment growth and outstanding employment growth in New York City—growth that, in fact, exceeded national levels over the past four years.
There is, of course, plenty of room for more improvement in the City tax climate. Even with the tax cuts of the last several years, New York remains by far the most heavily taxed big city in the country.
Our model underscores the important role that tax cuts can and should play in promoting continued growth of the New York City economy in years ahead.
NYC-STAMP is designed to identify the effects of tax policy changes by using standard, widely accepted statistical techniques and by creating a “model” of interactions between economic and tax variables, based on data gathered over a long period of time.
In technical terms, NYC-STAMP is a structural model in which the key variables (employment, the stock of capital, and wages) are expressed as a function of the relevant policy variables, notably City tax rates. Estimates of the reduced-form equations derived from the model were obtained from pooled time-series cross-section data for the time period 1975–1999. Estimation methods applied to these data corrected for common econometric problems arising from pooled, time-series, cross-section data.
The equations produced coefficients used to estimate the effect of tax rate changes on employment. The NYC-STAMP model is constructed to estimate this effect in two ways—as a change in the tax rate, or as a change in net revenue generated by the tax.
In the STAMP model, a change in tax law affects economic activity through the effects of that change on individual decisions to work, shop or create capital. NYC-STAMP shows how changes in the City tax rates on personal income, on sales, on property and on corporate profits affect jobs, payrolls, the City’s capital stock and tax revenues.
Putting The Model Through Its Paces
To illustrate how NYC-STAMP estimates the effects of the tax changes considered here, consider the effects of a hypothetical change in the City’s personal income tax. The income tax is imposed at graduated rates as shown in Table 1.
Table 1: NY City Tax Rates by Tax Bracket
Tax Brackets for Joint Filers
2002 Current Law
2002 Base rates, no surcharge
Up to $21,600
More than $90,000
The decline since 1998 reflects several changes—the expiration in 1999 of the 12.5 percent “Safe Streets, Safe Cities” surcharge first imposed in 1990, a state-subsidized STAR rate reduction of two-tenths of a percent off the statutory base rate, and the partial rollback of the 14 percent surcharge imposed in 1991. A portion of the latter surcharge remains, with an impact ranging from an additional 10.5 percent of the base rate in the highest bracket to 3.5 percent of the base rate in lower brackets, as can be understood by comparing the rates in the table above.
How would the City’s economy respond if all remnants of the surcharge were repealed? Table 1 shows what the 2002 tax rates would be under this scenario, compared to current law.
To predict the economic effects of this change, it is necessary to determine how it would affect the tax that an average New York worker pays on an additional dollar of income. This “average marginal tax rate” is important because it measures the penalty that applies to expanding work effort.
The tax rates shown in Table 1 are marginal rates applicable to individual brackets. By weighting these individual tax rates by the gross income received in each bracket, it is possible to arrive at the average marginal rate. The average marginal rate will be greater, the greater the tax rate applicable to each bracket and the greater the fraction of aggregate gross income that is concentrated in the upper brackets. The average marginal tax rate for 2002 under current law would be 3.35 percent. With the surtax removed, it would fall by .25 percentage point to 3.10 percent.
This reduction in the average marginal tax rate promotes expansion of the City economy by increasing the after-tax wage rate—that is, the wage rate received by persons subject to the City income tax, after they paid their income tax. With every increase in wage rates, work becomes marginally more attractive for City residents.
The resulting expansion in the number of people willing and able to work in New York City would reduce the cost to employers of hiring workers and induce them, in turn, to hire more workers. One job of STAMP is to determine how the number of jobs increases as tax rates go down. Using regression analysis, the Beacon Hill Institute found that the number of New York City jobs rises by 1.87 percent for every one percentage point by which the City reduces the average marginal tax rate.
We can thus calculate the number of jobs the City would create in 2002 by cutting the average marginal tax rate by .25 percentage points. From an assumed projected baseline of 3,193,000 jobs in 2002, without any cut in the tax rate, the number of jobs would increase by:
(1) DL2002 = .25 X .0187 X 3,193,000 = 14,927.
Payrolls would rise, too. Payrolls per person working in New York are expected to average $62,918 in 2002. Thus the expected change in payrolls is:
(2) Dpayrolls2002 = 14,927 X $62,918 = $939 million.
The tax cut would also spur New York firms to create new capital. The stock of fixed private nonresidential capital (consisting of factories, offices, factory and office equipment and other such components of the City’s private infrastructure) in New York City is expected to be $446.590 billion in 2002. Regression analysis shows that the capital stock rises by 1.45 percent for every 1-percentage-point cut in the average marginal tax rate. Thus the capital stock would rise by:
(3) DK2002 = .25 X .0145 X $446.590 billion = $1.619 billion.
Table 2 summarizes these effects. The example illustrates how even small tax changes can exert significant effects on the city economy. The number of new jobs would translate in approximately a .5-percent increase in the total number of jobs. A rise in payrolls of almost $1 billion and an increase in the capital stock of about $1.6 billion would represent a substantial injection of economic activity into the City.
Table 2: Economic Changes in 2002 Resulting from Removal of Surcharge
Number of Jobs
Continuing, incremental reductions in the income tax surcharge, such as the ones adopted over the past year, make working in the City more attractive in the same way that cutting interest rates just a few basis points makes home mortgages or business loans more attractive to borrowers. Every little bit counts in making a financial determination whether to buy a house or deciding where to work and invest. Furthermore, the psychological effects of a program of steadily reducing tax rates count for a great deal. By keeping up its tax-cutting momentum through repeal of the remaining income tax surcharge, the City would send out a stronger signal that it is serious about reversing the high-tax image that previously made the City unattractive to prospective workers and that hindered capital formation.
Removal of the surcharge would cause the city to lose some tax revenue. The immediate “static” loss can be computed by determining how removal of the surcharge would affect personal income tax revenues if its removal exerted no expansive effect. Personal income tax revenues in 2002 are projected to be $5.453 billion.
In order to determine the effect on tax revenues of the cut in marginal tax rates, it is necessary to determine how that cut affects the City’s “average tax rate,” taken here to be the amount of personal income tax revenue the City collects for every dollar of adjusted gross income earned by persons paying City taxes. Removing the surcharge would cause this average rate to fall from 2.404 percent to 2.226 percent. The static effect on tax revenue therefore equals:
(4) DTRS2002 = [(2.226/2.404)-1] X $5.453 billion = – $404 million.
Because the tax cut would exert the above-noted expansive effects on the economy, however, this would not be the ultimate, net loss in tax revenue. There would be an increase in certain revenues that would partially offset the static loss.
Consider first how the tax cut would affect the personal income tax base. Regression analysis shows that this tax base, defined as adjusted gross income, rises by 71.3¢ for every dollar increase in payrolls. Thus there would be a dynamic change in personal income tax collections of:
(5) DTRDI2002 = .713 X .02226 X $939 million = $15 million.
Because payrolls would rise, so would retail sales. Regression analysis shows that the retail sales tax base rises by 26.7¢ for every dollar rise in payrolls. The effective sales tax rate is 5.26 percent. Thus there would be a rise also in sales tax collections equal to:
(6) DTRDS2002 = .267 X .0526 X $939 million = $13 million.
Finally, there would be a dynamic increase in corporate profits taxes. Regression analysis shows that the corporate profits tax base rises by 11.3¢ for every dollar increase in the capital stock. The average corporate profits tax rate is 5.29 percent. Thus corporate profits tax collections would rise by:
(7) DTRDC2002 = .113 X .0529 X $1.614 billion = $10 million.
The total dynamic effect is the sum of the increase in personal income tax, sales tax and corporate tax collections:
(8) DTRD2002 = $15 million + $13 million + $10 million = $38 million.
Subtracting the dynamic gain from the static loss, we arrive at the net revenue loss, which would be $366 million. The static, dynamic and net effects are summarized in Table 2. In effect, although this tax cut would have a budgeted “cost” of $404 million, using standard static analysis favored in official estimates, the dynamic feature of the model indicates that its net impact would be $38 million less.
1. The STAMP model was developed by the Beacon Hill Institute at Suffolk University in Boston, Massachusetts. BHI adopted the STAMP to the New York tax base, as further detailed in this report.
2. City of New York, Executive Budget, Fiscal Year 2002, Message of the Mayor, p. 17. The total includes the state’s 1999 repeal—over City objections—of the commuter wage tax.
3. “Report by the Chief Economist, Comptroller’s Budget Office, on the Impact of the Local Tax Burden on New York City,” Stephen Kagann, Ph.D, Chief Economist and Zheng Gu, Ph.D, Senior Economist, April 1991.
4. The state and City income tax cuts phased in between 1987 and 1990 did not actually produce as steep a tax reduction as the chart would indicate, because the lower rates applied to a taxable income base that had been considerably broadened by federal reforms.
5. Andrew Haughwout, Robert Inman, Steven Craig, and Thomas Luce, “Local Revenue Hills: A General Equilibrium Specification With Evidence From Four US Cities,” NBER Working Paper 7603, March 2000.
6. This figure is calculated using the equations described in Table 5 of the report by Haughwout et al., applying the methodology described in footnote 28 on page 33 of that report.
7. Estimate as described in comments by Prof. Robert P. Inman at Manhattan Institute Conference, “New York at the Crossroads: Visions of a Better City,” June 14, 2001.
8. Specifically, they measure whether New York and the three other cities are approaching the peak of their predicted revenue “hills.”
9. Data limitations prevented us from immediately developing a broad-based model that could also measure effects of reductions in more targeted taxes, including the hotel tax, commercial rent tax, unincorporated business tax and commuter tax.
10. The model does not estimate effects of reductions in the commercial rent tax, unincorporated business tax and commuter tax.
11. Due to rounding, the sum of these components does not precisely equal the total.
12. Federal job data for the City is available as far back as 1950.
13. Job data for June of each year, including 2001 estimated amounts, is from the U.S. Labor Department, Bureau of Labor Statistics.
14. The City managed to outperform the nation’s job trend for three consecutive years during the stagflation/recession period of 1980-82, when the worst job losses were concentrated in the manufacturing cities of the old Rust Belt. New York also managed to exceed the U.S. private job growth percentage during national slowdowns in 1954, 1958 and 1961.
15. Without steps to improve the efficiency and productivity of City services, the revenue loss associated with repealing the surcharge could also have negative consequences, if vital services are reduced. But even in this case, the change in the tax rate would in and of itself have an expansive effect.
16. Including, of course, persons who live outside the City and are not subject to the resident income tax.
17. This is not to be confused with the statutory rate. The effective sales tax rate is computed by dividing sales tax revenues by total retail sales.”