Starve the Beast: A Further Examination

TLDR: In this paper, Niskanen et al. found that reductions in federal revenues constrain neither the growth of peacetime federal spending nor the growth in nondefense discretionary spending.

Abstract: In recent years, some fiscal conservatives have argued that reducing tax revenues and increasing budget deficits is an effective strategy for limiting federal spending This strategy is commonly known as "starve the beast" Niskanen (2006) convincingly demonstrates that reductions in federal revenue do not limit the growth of federal expenditures Instead, he finds statistically significant evidence that revenue reductions actually stimulate the growth of federal spending However, proponents of starve the beast argue that low federal revenues might be able to limit the growth of certain components of the budget, such as nondefense discretionary spending Similarly, others argue that federal revenue reductions might be more effective at limiting expenditure growth during times of peace However, my analysis strengthens Niskanen's original research by finding that his results are consistent across time Furthermore, I find that reductions in federal revenues constrain neither the growth of peacetime federal spending nor the growth of nondefense discretionary spending Background Starve the beast argues that reducing tax revenues is an effective strategy for reducing, or at least limiting, government expenditures Specifically, proponents of starve the beast argue that low revenues, and the resulting deficits, will give elected officials the incentive to cut spending While a number of conservative activists have frequently used starve the beast as a justification for tax reductions, this theory has received support from some economists The most influential academic proponent of starve the beast is Milton Friedman Friedman (2003) argued that, if taxes are cut, "the resulting deficits will be an effective restraint on the spending propensities of the executive branch and the legislature" Other leading economists who have voiced support for starve the beast include Harvard University's Robert Barro (2001) who argued, "Tax cuts remove tax revenues from Washington and keep Congress from spending them" The first mention of starve the beast as it relates to the federal budget was in a 1985 Wall Street Journal article where an unnamed White House official felt that the Reagan administration had not done enough to cut spending: "We did not starve the beast," the official said (Blustein 1985) However, the ideas behind starve the beast have had some currency in mainstream political discourse since the late 1970s For instance, columnist George Will (1978) supported the enactment of the Kemp Roth tax reduction bill in 1978 because he thought "it would restrain the predictable growth of government that is financed by windfall revenues" Similarly, during the 1980 presidential debates Ronald Reagan argued that tax reductions would stop spending growth saying, "If you've got a kid that's extravagant, you can lecture him all you want to about his extravagance Or you can cut his allowance and achieve the same end much quicker" (Mallaby 2006) Much of the analysis of starve the beast has been largely anecdotal Some observers have argued that budget deficits in the 1980s helped President Reagan reduce the growth of nondefense discretionary spending Additionally, during the early years of the Clinton administration, some analysts argued that the Reagan-era deficits hindered President Clinton's efforts to increase expenditures on various programs (Edsall 1993) Furthermore, some observers have argued that the income tax rate reductions that President Bush signed in 2001 were an effort to put Congress in a "spending straightjacket" (Kinsley 2004) As such, even though starve the beast has provided a justification for those who wish to promote tax cuts, that theory has been subject to relatively little empirical scrutiny However, in 2006, William Niskanen, a former member of President Reagan's Council of Economic Advisers, ran a regression where he analyzed federal spending as a percentage of GDP from 1981 to 2005 …