Does “Crowd Out” Offset The Stimulus Effect Of Government Deficits? A Large Scale Econometric Study
John Heim ()
Rensselaer Working Papers in Economics from Rensselaer Polytechnic Institute, Department of Economics
This large scale econometric study finds private borrowing and spending decline as government deficits grow, due to “crowd out” effect resulting from financing the deficits from the limited pool of available loanable funds, and crowd out completely offsets stimulus effects. This result is found even controlling for business cycle effects, which can cause the same negatively correlated behavior. Separate tests of different types of tax cuts and government spending programs yielded the same result, as did tests of recession versus nonrecession periods. The models, tested on 50 years of data, explain very well the behavior of consumption and investment during the 2007-09 economic crisis, and simulation suggests that, ceteris paribus, stimulus programs of the type and composition of the 2009 Obama stimulus program would have a substantial negative effect on the economy, raising unemployment 1.25% -2.25% during the period they were in force. Negative effects of crowd out on consumer and business borrowing were also found, consistent with the spending findings. This supporting the underlying theory of crowd out, which is that reduced private borrowing (due to crowd out) is responsible for the observed negative relationship between private spending and deficit growth. Several hypotheses proposed by Krugman and others to refute the notion of crowd out negates the stimulus effects of deficits are tested. None are supported by the data. Hypotheses by Gale and Orszag that government transfer spending and federal tax cuts have positive stimulus effects, though others types might not, are tested. The hypotheses are not supported by the data. Well defined structural models of the U. S. economy 1960-2010 are tested. Extensive tests for endogeneity, stationarity, heteroskedasticity, as well as tests for robustness over time, with respect to model specification, and with respect to econometric technique were undertaken. Testing was done in 1st differences, eliminating most nonstationarity and reducing multicollinearity by approximately half. Models explained 90 -95% of the yearly changes of consumption and Investment during the 50 year period. Results were robust for tests of different time periods, generally fitting the data as well for the 1950s and 60s as for the 2000-10 period, and periods in between. Results were also robust for moderate changes to structural models, different regression techniques (OLS, strong and weak instrument 2SLS), and use of different strong 2SLS instruments.
JEL-codes: C50 C51 E12 E21 E22 (search for similar items in EconPapers)
New Economics Papers: this item is included in nep-fdg and nep-pbe
References: View references in EconPapers View complete reference list from CitEc
Citations Track citations by RSS feed
Downloads: (external link)
This item may be available elsewhere in EconPapers: Search for items with the same title.
Export reference: BibTeX
RIS (EndNote, ProCite, RefMan)
Persistent link: http://EconPapers.repec.org/RePEc:rpi:rpiwpe:1301
Access Statistics for this paper
More papers in Rensselaer Working Papers in Economics from Rensselaer Polytechnic Institute, Department of Economics Contact information at EDIRC.
Series data maintained by Shawn Kantor (). This e-mail address is bad, please contact .