One of the most lasting and costly pieces of economic destruction wrought by the crash of 2008 may be the extreme budgetary imbalances among the governments of high-income countries. The need for greatly expanded government outlays and declining tax collections now leaves governments scrambling to find viable methods of closing yawning public sector deficits.
Future public spending will be painfully curtailed without doubt, but feasible spending cuts are unlikely to restore fiscal balance. Painful tax increases will also be necessary.
In this environment, business taxes – more specifically, corporate income taxes – are natural targets for legislative tax-raisers. Business taxes share the expedient feature of appearing to be paid not by voters but by businesses, including corporations whose actions were at the centre of the 2008 troubles.
Business doesn’t pay business taxes: Shareholders, workers, or consumers do
As a theoretical matter, it has long been understood that it is nonsensical to say that businesses bear tax burdens (see, for example, Seligman 1899). The burdens created by business taxes must be borne by individuals, including shareholders and other owners in the form of reduced after-tax returns, workers in the form of lower wages, and customers in the form of higher prices.
Atkinson and Stiglitz (1980) drew attention to the possibility that using Harberger’s (1962) celebrated tax incidence model, corporate income tax burdens might be borne more than 100% by workers, whose wages could fall so much in response to the reallocation of resources triggered by the corporate tax (reduced corporate output and greater non-corporate output) that owners of corporations actually could come out ahead. So the identities of those who bear the burdens of business taxes depend very much on economic circumstances. These circumstances are potentially discoverable with empirical analysis.
Do higher corporate taxes cause lower wages?
The available evidence suggests that wages fall in response to higher corporate tax rates, implying that business tax burdens are largely shared with workers.1 One of the great difficulties in this line of research lies in identifying what wages would have been in the absence of high corporate tax rates. Wages are functions of local conditions and worker characteristics that may be unobserved by the analyst and, much worse, correlated with local tax rates, creating the prospect of attributing effects to taxation that properly belong to the local conditions that are correlated with taxation.
Our recent study (Felix and Hines 2009) largely sidesteps this issue by considering the effect of US corporate taxes on union wage premiums, which measure the extent to which union wages exceed wages of comparable non-unionised workers. Union wage premiums reflect a combination of employer profitability and the ability of unions to share in those profits. High corporate tax rates reduce firm profitability and thereby reduce the available pool of resources that unions can command. By comparing the effects of taxation on unionised and non-unionised workers, it is possible to abstract from economic factors that are common to both and instead focus on the extent to which tax burdens are shared with unionised workforces.
The US evidence exploits differences between rates of corporate taxation in US states, which range from 0% to 12%. Comparing workers in states with tax rates above 9% to those in states with tax rates below 4%, the evidence is that union wage premiums are $1.88/hour higher in the low-tax states than in the high-tax states. Regressions that use the whole sample of US workers from every state (collected in the 2000 Current Population Survey), and that control for education, age, race, marital status, other worker characteristics, and other local characteristics, produce a similar pattern – a 1% lower state corporate tax rate (e.g., the difference between a six and seven% tax rate) is associated with a 0.36% increase in union wages. The estimates imply that for a corporation whose workforce is fully unionised, any state tax saving it may enjoy is shared roughly half with its union. Since firms are typically not fully unionised, average effects are rather smaller than this.
It is always possible that this pattern appears by chance, that union wage premiums simply happen to be higher in high-tax locations for reasons unrelated to employer tax burdens. There are ways to refine the statistical test in order to guard, at least in part, against this possibility.
Our first method compares workers in highly capital-intensive industries with those in less capital-intensive industries. Corporate taxes are essentially a tax on capital, and therefore capital-intensive firms should be the most hard-hit by high corporate tax rates. In addition, relatively capital-intensive firms receive less help from the normal process by which corporations pass the burden of taxes onto workers (i.e. as competitive wages fall in response to higher corporate tax rates, and thereby lessen the burden on investors, this has a smaller impact on the bottom lines of capital-intensive firms). Consequently, any tax effects on union wage premiums should be most pronounced in capital-intensive industries – which, according to the US data, they are. The evidence is that state taxes have significantly greater effects on the wages of workers in capital-intensive industries than they do on wages of workers in less capital-intensive industries.
A second refinement of the statistical test lies in comparing locations in which unions are more powerful to those in which unions are less powerful. Roughly half of American states have enacted “right-to-work” laws that permit non-unionised workers to receive the same benefits and employment opportunities as unionised workers. “Right-to-work” laws are commonly, and accurately, thought to reduce union power by creating a more pronounced collective action problem among workers who otherwise would be inclined to see strong benefits from active union bargaining. The evidence is that virtually all of the effect of corporate taxes on union wage premiums appears in states without “right-to-work” laws, where unions are more powerful.
The evidence that US union wage premiums decline at higher rates of taxation serves as a reminder that stakeholders of all kinds – workers, suppliers, customers, executives, and certainly owners – share in the rents produced by profitable firms. Efforts to tax firms more heavily create burdens that will be distributed among stakeholders, possibly including many groups that governments otherwise attempt to help.
1 See, for example, Hassett and Mathur (2006), Felix (2007), Arulapalam, Deveruex and Maffini (2007), Desai, Foley and Hines (2007) and Felix (2009); Auerbach (2006) and Gentry (2007) provide critical surveys of the empirical corporate tax incidence literature.
Arulampalam, Wiji, Michael P. Devereux, and Giorgia Maffini (2007), “The Direct Incidence of Corporate Income Tax on Wages,” Working Paper 07/07, Oxford University Centre for Business Taxation, July.
Atkinson, Anthony B., and Joseph E. Stiglitz (1980), Lectures on Public Economics. New York: McGraw-Hill.
Auerbach, Alan J. (2006), “Who Bears the Corporate Tax? A Review of What We Know,” in James M. Poterba, ed., Tax Policy and the Economy, vol. 20. Cambridge, MA: MIT Press.
Desai, Mihir A., C. Fritz Foley, and James R. Hines Jr. (2007), “Labour and Capital Shares of the Corporate Tax Burden: International Evidence,” Working Paper, Harvard University, December.
Felix, R. Alison (2007), “Passing the Burden: Corporate Tax Incidence in Open Economies,” Working Paper, Federal Reserve Bank of Kansas City, October.
Felix, R. Alison (2009), “Do State Corporate Income Taxes Reduce Wages?,” Federal Reserve Bank of Kansas City Economic Review, vol. 94, no. 1, pp. 5-30.
Felix, R. Alison and James R. Hines Jr. (2009), “Corporate Taxes and Union Wages in the US,” NBER Working Paper 15263, August.
Gentry, William M. (2007), “A Review of the Evidence on the Incidence of the Corporate Income Tax,” Office of Tax Analysis, OTA Paper 101, December.
Harberger, Arnold (1962), “The Incidence of the Corporation Income Tax,” Journal of Political Economy, vol. 70, no. 3, pp. 215-240.
Hassett, Kevin A. and Aparna Mathur (2006), “Taxes and Wages,” Working Paper #128, American Enterprise Institute for Public Policy Research, June.
Seligman, Edwin R. A. (1899), The Shifting and Incidence of Taxation, Second Edition. New York: Columbia University Press.