Crowding Out Fiscal Stimulus Testing the Effectiveness of US Government Stimulus Programs John J Heim www.ebook3000.com Crowding Out Fiscal Stimulus John J Heim Crowding Out Fiscal Stimulus Testing the Effectiveness of US Government Stimulus Programs www.ebook3000.com John J Heim University at Albany-SUNY Albany, New York USA ISBN 978-3-319-45966-0 ISBN 978-3-319-45967-7 (eBook) DOI 10.1007/978-3-319-45967-7 Library of Congress Control Number: 2016954261 © The Editor(s) (if applicable) and The Author(s) 2017 This work is subject to copyright All rights are solely and exclusively licensed by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed The use of general descriptive names, registered names, trademarks, service marks, etc in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication Neither the publisher nor the authors or the editors give a warranty, express or implied, with respect to the material contained herein or for any errors or omissions that may have been made Cover illustration: © Everett Collection Historical / Alamy Printed on acid-free paper This Palgrave Macmillan imprint is published by Springer Nature The registered company is Springer International Publishing AG The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland For Sarah and Tom J.B and Jenn Lindsay and Luke Kelly and Casey Scott and Ross www.ebook3000.com EXECUTIVE SUMMARY Do deficit-financed government fiscal stimulus programs actually stimulate the economy? This study exhaustively tests a wide variety of different stimulus models, testing them in different time periods, and using different regression techniques in an attempt to answer this question Fiscal stimulus programs examined include both those that cut taxes and those that increase government spending Most models that predict government deficits will stimulate the economy are Keynesian A principal characteristic is that they are demand-driven Each of their key structural equations indicate increases in demand will lead to increased supply (and employment), at least up to full employment of resources Hence a deficit, which increases government demand by increasing government spending, or increases private demand by cutting taxes, stimulates the economy, according to stimulus theory In testing whether these stimulus programs actually work, we took care to test equations taken from the Keynesian model This gives the stimulus effects of deficits, should they exist, the best possible chance of being verified empirically 1960–2010 data on the US economy were used A total of 228 models of the determinants of consumption, investment and the GDP were tested These models differ in the variables included, the time period tested, the regression techniques used, the level of the economy when testing and the specific type of tax cut or spending stimulus used Each model is reported in detail, with differences in each subsequent model compared to the last tested Differences in results are compared as well The work is lengthy, but necessarily so Our goal was to test every vii viii EXECUTIVE SUMMARY conceivable model readers might feel stood a chance of showing positive stimulus effects That way, whatever our results turned out to be, it would be difficult to argue we did not examine a wide enough breath of stimulus programs to give them a fair chance to show what they can In theory, the effectiveness of stimulus programs can be curtailed by “crowd out” Crowd out theory suggests that whatever their stimulus effects, government deficits have the undesirable result of simultaneously reducing private spending, because funds normally borrowed by consumers and businesses must be used to fund the deficit Reduced private borrowing in turn causes a reduction in consumer and business spending, offsetting the deficit’s stimulus effects In testing to show crowd out effects, a deficit variable is added to each Keynesian consumption, investment or GDP model This allows simultaneous testing of both Keynesian stimulus and crowd out effects Each model shows stimulus and crowd out effects separately, and allows readers to directly calculate net effects of stimulus programs, for example, the effects of reductions in federal taxes on consumption spending, controlling for many other factors that affect consumer spending Everybody acknowledges the possibility that crowd out, “if it occurs”, can adversely affect stimulus programs Rarely people actually test to see if it does occur This book’s unique contribution is to exhaustively test to determine if it does occur, and if so, how serious a problem it is As alternatives to Keynesian models, we could have tested dynamic stochastic general equilibrium (DSGE) models, but these are models that typically contain assumptions about human behavior like perfect foresight and intertemporal utility maximization, that are designed to infer from these assumed behavioral characteristics that stimulus programs don’t work in the long run Similarly, we could have picked different combinations of five or six variables thought appropriate and tested VAR models The problem here is that since VARs typically are not recognizable theoretical constructs, it is hard to know what your results mean By comparison, we know what a Keynesian theoretical construct means, and if we slip a deficit variable into it and then test, the test will either show the variable statistically significant (crowd out matters) or not (crowd out doesn’t matter), controlling for all the other Keynesian influences In short, we will have a useful result, from a fair test, scientifically arrived at When examining how consumption or investment spending varies as deficits rise or fall, we control for the effect of business cycle variation on both government deficits and on private spending This was necessary to www.ebook3000.com EXECUTIVE SUMMARY ix unambiguously identify crowd out’s effects A declining economy alone can cause a growing deficit and simultaneously declining consumer and business spending, but that is a business cycle effect, not a crowd out effect in which the deficit causes the decline in private spending Some argue it is only business cycle effects, not crowd out, which cause the negative relationship between deficits and private spending They argue that without the stimulus programs, the observed decline in private spending associated with the deficit would have just been worse The difference between this and the crowd out explanation has enormous political and economic implications for government’s role in the economy Methodologically, extensive tests for endogeneity, stationarity and heteroskedasticity were undertaken Testing was done in first differences, eliminating most non-stationarity and reducing multicollinearity problems by about half Models explained 90–95 % of the yearly changes of consumption and investment during the 50-year period tested Four different, though overlapping, time periods were tested Findings were essentially the same for the 1950s and 1960s as for the 2001–10 decade, and for decades in between Deficit results were also robust for moderate changes in the structure of the models tested They were also generally robust to different regression techniques (OLS, strong and weak instrument 2SLS), and use of different strong 2SLS instruments Hence, we feel our findings will be difficult to refute in reasonably well-constructed future models of how the Keynesian system works Our findings overwhelmingly indicate that as government deficits grow, creating observable stimulus effects, consumer and business spending declines due to crowd out, fully cancelling the stimulus effects (or worse) The offsetting decline in private spending appears due to “crowd out”, that is, trying to finance both increased deficits and traditional private borrowing levels from a relatively unchanging sized pool of loanable funds This finding that stimulus programs not work held in virtually all circumstances tested The specific type of tax cut or government spending deficit did not matter, nor did it matter if we tested in recessions or normal economic times Nor did it matter what particular decades since 1960 we tested, and, generally, it did not matter if we used one regression technique (OLS) versus another (2SLS) to our analysis (though as a matter of good practice, where 2SLS is needed, it should be used) With rare exceptions, usually due to statistical problems like multicollinearity, none of these variations in the models we tested resulted in a net positive effect for tax x EXECUTIVE SUMMARY cut or spending increase stimulus programs In virtually all cases, results indicated crowd out fully, or more than fully, offset stimulus effects Examination of the 1981–83 recession period indicates the pool of loanable funds drops even faster in recessions than business and consumer loan demand Hence, new deficit financing demands on the pool of loanable funds in recessions, if anything, cause even bigger crowd out problems than in normal times The data examined support this conclusion The models from which we obtained these results also explain extremely well the behavior of consumption and investment during the 2007–09 economic crisis Our econometric findings suggest that deficit-financed stimulus programs such as the 2009 Obama stimulus program have a substantial negative effect on the GDP, raising unemployment 2.26– 2.94 % during the period they are in force Deficits have this undesirable result because to fund them, funds normally borrowed by consumers and businesses are used Reduced borrowing by consumers and businesses in turn causes a reduction in their spending, offsetting the deficit’s stimulus effects Worse, “lumpiness” in borrowing tends to result in private spending reductions even greater than the stimulus’ positive effects, leading stimulus programs to have a net negative impact on the economy By lumpiness we mean the following: consumers who need to borrow $10,000 to buy a new car, but find that their bank can only lend them $9000 (because they lent the other $1000 to the government to finance a $1000 stimulus program), will not buy the car at all This causes private spending to fall $10,000 from expected levels, a far greater drop than the stimulus can offset Chapter 16 provides a more detailed summary of findings and conclusions www.ebook3000.com PREFACE I left academic life in 1972, not to return until a quarter century later When I returned, one of the most hotly contested issues of my youth, “Do Keynesian-type stimulus programs work?” was still unresolved I was surprised because when I left academia, work in economics seemed more and more dominated by the new, econometrically based scientific method, rather than the older philosophical approach, i.e., mainly theoretical deductions derived from “self evident” truths about human and business behavior I felt it would only be a matter of time before science provided an answer to the stimulus question That did not occur My research interests in large-scale econometric modeling led me to try to develop and test a Keynesian-type model of the macroeconomy For about six months, I kept trying to build and test simple Keynesian Cross and IS-LM models, and then extend the work to more complex models of the same type, but with no success In empirical test after test, I kept coming up with the wrong sign on the government revenues variable: I was consistently getting positive signs, when Keynesian stimulus theory said I should be getting negative signs Worse, I was having nearly as bad a problem with my government spending results In more sophisticated models, test results for government spending were also giving me the wrong sign: negative instead of the positive sign Keynesian theory leads us to expect What to do? One thought was just scrap my Keynesian model testing program and move on to testing some other theory This is clearly what many of my colleagues had done during the 1980s and 1990s when I was out of macroeconomics What large-scale models remained were now DSGE-based xi 256 J.J HEIM Table 6.6 Effect of deficits on investment spending (Per dollar of deficit) (Per dollar of borrowing) Spending tests Deficit Effect Borrowing Effect Average (All Tests) Av.(2SLS Str.Inst.Only) Average (OLS) $ 0.30 (3.6) to 0.35 (5.5) $ 0.30 (3.4) to 0.34 (3.9) $ 0.30 (3.6) to 0.33 (5.5) 0.09 (1.8) to 0.19 (1.7) 0.09 (1.8) to 0.19 (1.7) 0.10 (1.9) to 0.13 (2.1) t-statistic in parentheses Table 8.4 Effect of deficits on investment spending (Per dollar of deficit) (Per dollar of borrowing) Spending tests Tax Def Spend Def Borrowing effect Average (All tests) Av.(2SLS Str.Inst.Only)1 Average (OLS) $0.29 (4.0) $0.23 (2.9) $0.27 (2.8) À0.56 (À4.9) 0.07(0.4) À0.47 (À4.1) 0.11(1.9) À0.40 (3.5) 0.10 (1.9) t-statistic in parentheses Models using ProfÀ0, DJ0 for, indicating crowd out’s effect was not being confused with Krugman’s noted business cycle effect, but was something that occurred separately Detailed presentation of individual test statistics are provided and summarized at the end of Chaps 5, 6, and Chapter compares the and variable deficit results and finds them very consistent for tax deficits, and roughly consistent for spending deficits For two-variable deficits, estimates of the negative effects of spending deficits are noticeably greater Chapter 10 (Section 10.1) estimates the effects of government stimulus programs, net of crowd out, on the GDP, rather than just consumption or investment alone Two different methods of testing are used Both indicate deficits induced by tax cuts had a decidedly large, statistically significant net negative effect on the GDP ($0.47–$0.89 per dollar of deficit), and spending deficits had a smaller, but still net negative effect on the GDP ($0.35– $0.57 per dollar of deficit) over the past 50 years, but the spending effect was, statistically speaking, insignificantly different from zero The more negative effect of tax cut deficits stems from their smaller initial stimulus SUMMARY OF FINDINGS AND CONCLUSIONS 257 effect, which acts initially through the marginal propensity to consume (less than 1.0 per dollar of tax cut), compared to the initial stimulus effect of government spending (1.0 per dollar of deficit) Both are later magnified through multiplier effects Unlike stimulus effects, crowd out effects tend to be the same, more than offsetting initial smaller stimulus effects of tax cuts, just offsetting the larger stimulus effects of spending projects Chapter 10.2 evaluates the four Krugman hypotheses in light of these findings The analysis concludes as follows: (1) Crowd out, not changes in the business cycle, is the major cause of the observed decline in private consumer and investment spending when deficits rise (2) Tests indicate net stimulus program effects are always zero or negative, regardless of stimulus size Hence, any observed failure of stimulus programs to work is not because they are too small, or because their positive effects were used up just stopping the economy’s crash (3) The crowd out effect of deficits is as great in recessions as normal economic times This is because savings (the pool of domestic loanable funds) drops as much as loan demand during recessions This leaves the reduced pool just equal to the reduced loan demand Hence, any government borrowing in recessions creates the same crowd out problem, leaving stimulus programs as ineffective In some cases, the pool of loanable funds may drop faster than loan demand, causing an even worse situation (4) Deficit-financed stimulus programs not stimulate investment during recessions by improving the economy, causing business confidence to increase, and “crowding in” to result Test results indicated their effect on the economy is typically negative, creating just the opposite effect Chapter 10 (Sect 10.3) tests for Gale and Orszag effects A 2004 study by former federal budget director Peter Orszag and economist William Gale found that Federal tax cuts have a positive stimulus effect on the economy, though state and local tax cuts have a negative effect They also found that government spending on transfer income programs had a net stimulus effect on the economy, though spending on all other government programs did not The portions of the Gale and Orszag study dealing with state and local tax cuts and non-transfer government spending are consistent with this www.ebook3000.com 258 J.J HEIM study However, the findings regarding federal tax cuts and transfer spending are not Unlike Gale and Orszag, we did not find positive stimulus effects for federal tax cuts or for transfer spending increases Using the same models this study previously tested in Chaps 5, 6, 7, and 8, but dividing the two total tax and total government spending variables into Gale and Orszag’s four types of taxes and government spending, and retesting, we obtained no evidence of the Gale and Orszag effect We have noted earlier (See Section 2.3.2) that Orszag’s testing gave a net effect of tax cuts on consumer spending only, and did not factor in investment crowd out effects (as our IS curve models do) This may be a cause of the difference in results Chapter 10 (Section 10.4) uses the estimates of both the stimulative and crowd out effects to estimate the effects of the $830 billion 2009 Obama stimulus program The results indicated the program’s impact on the GDP was negative ($À340.3 billion to $À605.4 billion), a drop associated with an increase in the unemployment rate of 1.26–2.26 % The effect in 2010 would have been to raise the unemployment rate from 9.6 % to 12.2–13.2 %, had other factors that influence the GDP not been moving in a more positive direction and offsetting crowd out effects (exports, the exchange rate, profits, the effect of prior year declines in interest rates, etc These positive effects can also be estimated from the tests enclosed) On a more technical note, Chap 11 shows that while this study is done using comparative statics models, the same models could be relatively easily converted to dynamic models for further testing Chapter 12 notes that the crowd out problem can be avoided if borrowing is done from foreign sources of funding, provided these funds are not already part of the pool of foreign funds commonly used by domestic private borrowers Evidence from the US recession period of 1981–83, shows that the pool of loanable funds (savings) declined more than loan demand Hence, any government borrowing to finance a stimulus program during this recessionary period should have created a crowd out problem However, a large increase in foreign borrowing was able to avoid crowd out problems that would have otherwise occurred from the large growth in the government deficit during this period Chapter 13 also deals with a technical issue It shows that the definition of disposable income used in this study’s consumption functions (Y À T) is equivalent or better than the disposable personal income definition more commonly used in economic studies SUMMARY OF FINDINGS AND CONCLUSIONS 259 Chapter 14 tests 13 models to see if crowd out is as big a problem in recessions as in non-recession periods It finds crowd out is as much of a problem in recessions as in normal economic times This appears to be because the pool of loanable funds drops as much or more during recessions than loan demand, leaving any borrowing of part of it by government to fund a deficit as much a problem for private borrowers as during normal times The pool drops because the decline in income in recessions causes savings to drop This also was an important finding, since it is sometimes argued that even if crowd out is a problem normally, it is not in recessions because private borrowing declines, leaving funds available for government to borrow without hurting private borrowers Overall, Chap 14 indicates tax cut deficits have a negative effect on GDP in both recession and non-recession periods, and spending deficits have no effect on the GDP either way Chapter 15 extends the recession/non-recession effect analysis of Chap 14 to test for Gale and Orszag effects separately in recession and non-recession periods, testing ten additional models No evidence of Gale and Orszag effects was found in non-recession periods In recessions, state and local tax cuts and non-transfer spending were found negatively related to the GDP, and federal tax cuts and transfer spending were found to have no net effect Some initial evidence of positive transfer spending effects in recession periods was found, but was later evaluated as due to a statistical problem (multicollinearity), which when corrected, left no statistically significant evidence of a positive effect 16.1 CONCLUDING OBSERVATION The models tested in this study hypothesize a demand-driven economy Supply is assumed to be sufficiently elastic to meet changes in demand in relatively short periods of time, and consumer spending is assumed driven principally by current income (essentially a Keynesian economy), but other factors are important, too Testing of these models (when crowd out is included) indicated they explain 90–97 % of the yearly changes in consumer and investment spending, and the GDP, over the past 50 years Even without the crowd out variables, they explain much of the variation Earlier studies, referenced in the literature review section, indicate crowd out adds 6.4–15 % to explanatory power This is strong evidence that the hypotheses tested, and nitty-gritty details about their underlying structure, are reasonably accurate representations of how the economy operates www.ebook3000.com 260 J.J HEIM That said, typical Keynesian models fail to show that deficit-financed “stimulus” programs are likely to have a zero or negative effect on the economy, or explain why that might happen This results from errors of omission in the theory Some effects of deficits are stimulative, as the Keynesian’s point out (and our tests show them) However, there are some effects of deficits that can negatively influence the deficit’s total effect, and these are omitted in most tests of the standard Keynesian system These are the crowd out variables, the variables constituting the government deficit Including them is necessary to ensure there are variables in empirical tests which can capture the negative effects of government borrowing on private spending In this study, our fix for the standard system is not to scrap it, but just to add crowd out variables to it and retest The results are used to re-estimate the whole system, and the re-estimated parameters are checked to see if they are more consistent with observed reality We conclude they are Tests consistently indicate crowd out, along with disposable income and the investment accelerator, are the most important variables in the consumption and investment models The expanded system, including the crowd out variable, explains virtually all of the variation in consumer and investment spending that has occurred the past 50 years, as shown in the consumption and investment graphs below Hence, we conclude that not only did we successfully test for crowd out’s effect on stimulus programs, but we picked the best type of models to so—those that were already very good at explaining what drives economic behavior Billion (2005 $) 400 Change Estimated By Model 300 200 100 –100 Actual Yearly Change In Spending –200 1960 1965 1970 1975 Consumption spending 1980 1985 1990 1995 2000 2005 2010 SUMMARY OF FINDINGS AND CONCLUSIONS 261 Consumption Model Shown: ΔCT ¼ 0:50ΔðY À T T ị ỵ 0:55TT ị 0:26GT&I ị tẳị 11:4ị 11:4ị 3:7ị 11:81PR ỵ 0:42 DJ2 ỵ 3:42 XRAV 336:65POP16 5:1ị 5:3ị 2:3ị 1:3ị ỵ 0:012POP ỵ 0:36ICC1 ỵ 40:86M2AV 2:6ị 1:3ị 3:8ị ỵ 0:12 CB2 ỵ 0:04 GDPReal3ị 3:1ị 1:1ị R2 ẳ 94:9 % D:W: ẳ 1:8 MSE ẳ 25:45 7:1ị Billion (2005 $) 400 Actual Yearly Change In Spending 200 –200 –400 Change Estimated By Model –600 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 Investment spending Investment Model Shown: ΔI T ẳ ỵ 0:33ACCị ỵ 0:22TT ị 0:53GT&I ị tẳị 4:9ị 2:0ị 3:4ị ỵ 0:81DEP ỵ 2:39CAP1 2:29PR2 3:0ị 1:0ị 0:9ị ỵ 0:10DJ0 ỵ 0:13PROF0 ỵ 5:87XRAV 0:4ị 1:9ị 2:4ị ỵ 0:013POP ỵ 0:05 BOR1 ị 12:40 UNEM 2:8ị 0:9ị R ẳ 93:1 % D:W: ẳ 2:0 MSE ẳ 33:05 www.ebook3000.com 1:5ị 8:2:Alt:bị BIBLIOGRAPHY Barley, R 2010 The Hidden Risks Governments Create by Delaying Cuts Wall Street Journal, February 2, p.C14 Barro, R.J., and C.J Redlick 2011 Macroeconomic Effects from Government Purchases and Taxes Quarterly Journal of Economics 126(1): 51–102 Blanchard, O., and R Perotti 2002 An Empirical Characterization of the Dynamic Effects of Changes in Government Spending and Taxes on Output The Quarterly Journal of Economics 117(4): 1329–1368 Chan, S 2010 Group of 7’s Finance Ministers Stand By Stimulus Programs NY Times, February 2, p.A16 Conference Board, The 2003/2011 Economic Report of the President, 2002 and 2010 Washington, DC: U.S Government Printing Office Conference Board, The The Conference Board’s Index of Consumer Confidence The Conference Board, Inc 845 Third Avenue, New York, NY, pp 10022–6600 Eckstein, O 1983 The DRI Model of the U.S Economy New York: McGraw – Hill Ford, S 1986 A Beginner’s Guide To Vector Autoregression Staff Paper P86-28, Institute of Agriculture, Forestry and Home Economics University of Minnesota Fox, K 1968 Intermediate Economic Statistics New York: John Wiley and Sons Friedman, B 1978 Crowding Out or Crowding In? Economic Consequences of Financing Government Deficits Brookings Papers on Economic Activity 1978(3): 593–641 Furceri, D., and R Sousa 2009 The Impact of Government Spending on the Private Sector: Crowding Out versus Crowding In Effects NIPE WP 6/2009, Nucleo De Investigaca Politicas Economicas, Universidade Do Minho, Portugal ——— 2011 Flow of Funds Accounts of the United States, Data Series TODNS, TBSDODNS, CMDEBT, FGSDODNS, SLGSDODNS, AND DODFS © The Author(s) 2017 J.J Heim, Crowding Out Fiscal Stimulus, DOI 10.1007/978-3-319-45967-7 263 264 BIBLIOGRAPHY Washington, DC Board of Governors of the Federal Reserve System March 11, Update www.federalreserve.gov/releases/z1/current/ Gale, W.G., and P.R Orszag 2004 Budget Deficits, National Saving, and Interest Rates Brookings Papers on Economic Activity 2004(2): 101–187 Granger, C.W.J 1969 Investigating Causal Relations by Econometric Models and Cross-Spectral Methods Econometrica 37: 424–438 Hall, Robert 1978 Stochastic Implications of the Life Cycle—Permanent income Hypothesis: Theory and Evidence Journal of Political Economy 86(6): 971–987 Heim, John J 2007 How Much Does The Prime Interest Rate Affect U.S Investment? Journal of the Academy of Business and Economics 7(1): 143–149 ——— 2010 Do Government Deficits Crowd Out Consumer and Investment Spending? Journal of the Academy of Business and Economics 10(3) Also available by the same title as Working Paper # 1005, Department of Economics Working Paper Series, Rensselaer Polytechnic Institute, Troy, N.Y., July 2010 ——— 2011 Is Crowd Out A Problem In Recessions? Working Paper # 1102, Department of Economics Working Paper Series Troy, NY: Rensselaer Polytechnic Institute ——— 2012a Does Crowd Out Hamper Government Stimulus Programs In Recessions? Journal of Applied Business and Economics 13(2): 11–27 ——— 2012b The Different Crowd Out Effects of tax Cut and Spending Deficits Applied Econometrics and International Development 12(2): 105–122 Jaeger, A 1992 Does Consumption Take a Random Walk? The Review of Economics and Statistics 74(4): 607–614 Kraay, A 2012 How Large is the Government Spending Multiplier? Evidence From World Bank Lending Quarterly Journal of Economics 127: 829–887 Krugman, P 2009 Crowding In New York Times, September 28 ——— 2013 The Story of Our Time New York Times, April 28 Mankiw, G 2006 The Macroeconomist as Scientist and Engineer Journal of Economic Perspectives 20(4): 29–46 Molena, H 1991 The Time Series Consumption Function: Error Correction, Random Walk and the Steady-State The Economic Journal 101(406): 382–403 Mountford, A., and H Uhlig 2008 What Are The Effects Of Fiscal Policy Shocks? NBER Working Paper # 14551, December NBER 2016 S Business Cycle Expansions and Contractions National Bureau of Economic Research, 1050 Massachusetts Avenue, Cambridge, MA, 02138 Sims, C.A 1972 Money, Income and Causality The American Economic Review 62 (4): 540–552 (Sept 11972) ——— 1977 Macro-economics and Reality Center for Economic Research Discussion Paper No 77–91 Department of Economics, University of Minnesota, Minneapolis, Minnesota www.ebook3000.com BIBLIOGRAPHY 265 Solow, R.M 1956 A Contribution to the Theory of Economic Growth Quarterly Journal of Economics, 70(1): 65–94 ——— 2010 Testimony to the Subcommittee on Oversight and Investigations of the Committee on Science and Technology, U.S House of Representatives hearing on “Building a Science of Economics for the Real World” Spencer, R.W., and W.P Yohe 1970 The Crowding Out of Private Expenditures by Fiscal Policy Actions Federal Reserve Bank of St Louis Review Triola, M 2012 Elementary Statistics, 11 edn Boston: Addison Wesley Uhlig, H 2005 Journal of Monetary Economics 52: 381–419 Woodford, M 2011 Simple Analytics of the Government Expenditure Multiplier American Economic Journal: Macroeconomics 3(1): 1–35 INDEX A accelerator, 6, 8, 12, 28, 29, 33, 35, 63, 92, 94, 107, 176, 180, 181, 200–3, 211, 260 accommodating, 12, 27, 179, 180 Alt, 67, 68, 71, 75, 77–85, 108–21, 123, 124, 128, 129, 135, 137, 139–41, 143, 145–9, 152–60, 162–4, 166, 170, 177, 200, 218, 220, 232, 243, 261 Alt2, 64, 72, 75, 80–2, 84, 85, 110, 121, 138, 139, 141–6, 148, 149, 155 "a" versions, 78, 101, 102, 152 B Barley, R, 12 Barro and Redlick, 25–8 Blanchard and Perrotti, 23 borrowing business, 29, 34, 40, 41, 91–5, 97, 101–6, 108–14, 116–20, 122, 124–7, 130, 161–6, 177, 186, 209, 211, 221, 255 consumer, 23, 34, 39, 41, 43–7, 58–62, 65, 76–9, 83, 84, 86, 90, 91, 101, 102, 105, 134, 139, 140, 142, 144, 146–9, 151, 166, 211, 221, 233, 253 foreign, 28, 179, 209–11, 213, 214, 258 government, 1, 37, 58, 169, 179, 182, 213, 214, 234, 252, 254, 257, 258, 260 lumpy, 233 business confidence, 176, 184, 204–8, 257 business cycle, 2, 15, 34, 45, 92, 133, 151, 175, 229, 256 business cycle control, 15–17, 34, 37, 38, 45–57, 59–61, 65–74, 77–84, 87–9, 94–100, 102–5, 108–10, 112–18, 122, 125, 130, 131, 133–8, 140–5, 151–9, 162–4, 177, 185, 186, 188–90, 229, 230, 240, 242, 243, 245, 246 business indebtedness, 41 C capacity utilization, 28, 64, 92, 108 Chan, S., 11 © The Author(s) 2017 J.J Heim, Crowding Out Fiscal Stimulus, DOI 10.1007/978-3-319-45967-7 www.ebook3000.com 267 268 INDEX consumer confidence, 7, 28, 33, 44, 52, 64, 66, 69, 107, 108, 125, 135, 137, 138, 204, 205, 207, 208, 211, 221 consumer indebtedness, 41 consumer spending, 8, 13, 23, 29, 39, 41, 43–94, 111, 133–49, 179, 180, 204, 217, 220, 231, 233, 253–5, 258, 259 consumption, xii, 2, 5, 7, 8, 13, 15, 17–20, 22–6, 28–31, 33–6, 40, 41, 43–7, 50–3, 56, 58, 62–5, 73, 74, 77, 85, 87–9, 92, 93, 101, 107, 108, 110, 121, 124, 127, 131–3, 136–40, 143, 145, 148–9, 151, 161, 167, 173, 175–82, 188, 199, 201, 204, 211, 215–28, 231, 232, 251, 255, 256, 258, 260, 261 covariance, 17, 183, 184, 186, 236 crowd out, xii, 1, 5–11, 33, 43, 91, 133, 151, 173, 175–97, 209, 217, 219–37, 240, 251 crowd out effect, xii, 2, 3, 8, 9, 11, 12, 14–18, 23, 24, 27, 29, 34, 43, 44, 46, 51, 55, 56, 58–62, 70, 76–83, 86, 88, 91, 93, 97, 101–5, 115, 122–6, 130–2, 137–9, 144, 146–8, 161, 167, 169, 171, 174–97, 209–14, 218–37, 251–3, 255, 257, 258 D Data Resources, Inc., 11, 19 deficit, xii, 1, 11, 34, 43–149, 151–71, 173–5, 199, 209, 217, 219, 239, 251 deficit financed, 37, 60, 89, 167, 176, 182, 184, 187, 204, 206, 208, 211, 249, 257, 260 demand, 2, 3, 5–8, 12, 15, 16, 19, 28, 33, 34, 40–1, 45, 49–52, 54, 55, 76, 93, 94, 102, 122, 124, 131, 148, 161, 168, 180, 182, 187, 213, 219, 221, 225, 257–9 demand driven, 5, 7, 19, 45, 93, 131, 259 depreciation, 7, 28, 33, 63, 89, 90, 92, 107, 110, 113, 115, 117, 121, 212, 221 Dickey–Fuller, 15, 36, 220, 221 discontinuities, 9, 178, 179 disposable income, 5, 9, 14, 15, 18–20, 28, 29, 33, 35, 44, 60, 63, 68, 69, 72, 80, 82, 89, 107, 110, 113, 136, 137, 180, 201, 211, 215–18, 221, 234, 258, 260 dummy variable, 47, 206, 228 Durbin Watson, 29 dynamic effects, 199–208 E Eckstein, Otto, 1, 11, 12, 18, 25, 31, 251 econometricians, 1, 251 econometric modeling, xi, 2, 12, 19, 132, 214 econometrics, xi, 2, 6, 12, 19, 35, 36, 132, 195, 203, 213, 214, 237, 252, 253 Economic Report of the President, 32, 33, 212, 216, 221 endogeneity, 2, 6, 17, 18, 29, 35, 36, 46, 63–5, 71, 72, 77, 80–2, 84, 88–91, 106–8, 110, 111, 114, 117, 121, 124, 125, 127, 141, 161, 215, 216, 220, 222, 230, 244 endogeneity, Hausman, 35, 63, 71, 80, 88–90, 110, 111, 216 endogenous, 6, 18, 19, 31, 35, 36, 63–75, 77, 80, 82, 84–6, 89, 90, 107–10, 112–22, 125, 127, 134, INDEX 135, 137, 138, 140–5, 152–5, 158–60, 162–5, 182, 185, 203, 220, 222, 225–7, 229, 230, 246 error of commission, error of omission, 7, 260 Euler model, 18, 25, 31 exchange rate, 13, 28, 33, 44, 62, 64, 68, 77, 83, 92, 107, 144, 211, 221, 258 exogenous, 19, 31, 35, 36, 63, 64, 71, 72, 80, 84, 89, 107, 108, 110, 125, 137, 200, 203, 220 explanatory variables, 19–21, 26, 40, 45, 48, 50, 63, 64, 80, 89, 90, 93, 98, 108, 111, 112, 120, 122, 134, 161, 165, 199–201, 203, 204, 241, 243, 244 exports, 6, 9, 26, 40, 64, 77, 107, 108, 125, 135, 177, 182, 201, 258 F Fair, Ray, 19, 25, 27, 31 Federal reserve, 2, 33, 40, 41, 45, 59, 93, 221 federal revenue, 43, 216 federal taxes, 12, 13, 15, 17, 18, 43, 53, 57, 176, 188, 189, 191, 239–42, 244–6, 257–9 first differences, 14, 15, 17, 29, 36, 37, 50, 221 fiscal policy, xii, 1, 7, 11, 23, 132, 251 fiscal stimulus, 171, 175, 251, 252 flow of funds, 2, 33, 41, 45, 93, 187, 211–14, 221 Ford, S., 19, 20, 22, 31 Fox, Carl, 243 Freidman, Benjamin, 12 full employment, vii Furceri and Sousa, 24 269 G Gale and Orszag, 12–19, 31, 53, 56, 176, 188–91, 239–49, 257–9 GDP, 5, 14, 34, 44, 92, 93, 133, 151, 175–97, 199, 215, 220, 239, 252 goods and services, xii, 2, 7, 15, 18, 24, 35, 37, 43, 44, 47, 48, 51, 87, 122, 135, 148, 149, 161, 169, 171, 180, 182, 188, 210, 215, 217, 225, 232, 234, 253, 260 goods and services spending, 43, 188, 217 government debt, 11, 12 government purchases, 12 government spending, xi, xii, 3, 5–9, 12–15, 17, 18, 20, 22–8, 34, 39, 43, 53–8, 85, 91, 131, 133, 136, 137, 142, 146, 148, 161, 165, 168, 169, 175–8, 181–3, 185, 186, 188–90, 192–4, 196, 197, 200, 202, 211, 223–8, 234, 235, 237, 239, 240, 242, 244–6, 249, 252, 253, 257, 258 Granger causality, 20 H Hausman test, 6, 64, 72, 90, 107, 134, 135, 140–3, 145, 152–5, 158, 159, 162–4, 215, 220, 229, 230 Heim, John J., xii, 2, 25, 27–9, 31, 47, 148, 210, 211 heteroskedasticity, 37 Hill, Griffiths & Lim, 36, 37, 64, 182 I identification, 22, 31, 35 imports net exports, 6, 9, 26, 40, 177, 182, 201 www.ebook3000.com 270 INDEX impulse response, 20–2 instruments, 2, 6, 34, 44, 107, 135, 152, 173, 177, 217, 220, 244, 255 interest rate, 3, 6, 8, 11–13, 21, 28, 31, 33, 35, 44, 62, 64, 72, 75, 92, 148, 178, 211, 213, 214, 221, 258 interest rate, Prime, 28, 44, 62, 64, 72, 92 investment, 2, 5, 14, 33, 43, 91–132, 151–71, 173, 175, 199, 211, 219, 251 investment spending, 2, 8, 15, 16, 24, 29, 30, 33, 35, 38, 39, 41, 91–132, 151–71, 173, 177, 199, 220, 221, 229–31, 233, 253, 255–7, 259–61 IS curve, 6, 7, 17, 26, 40, 175–7, 182, 183, 185, 188, 189, 199–203, 220, 231–7, 239, 252, 258 IS–LM model, xi J Jaeger, A., 25, 27, 31 K Keynes, 26 Keynesian, xi, xii, 2, 5–10, 19, 20, 23, 27, 34, 40, 131, 132, 171, 175, 199, 204, 237, 252, 259, 260 Keynesian Cross, xi, 6, Keynesian mechanics, Keynesian stimulus, xi, 5–10, 20, 23, 252 Klein, Lawrence, 25 Kraay, A., 28 Krugman hypotheses, 184–8, 257 Krugman, Paul, 167, 253 Kuznets, Simon, 25, 215 L lagged GDP, 38, 39, 48–50, 55, 64, 68, 69, 87, 92, 113–15, 136, 142, 154, 188, 240, 243 loanable funds, 1, 2, 8, 9, 12, 39, 40, 48, 50, 58, 60, 65, 76, 86, 93, 122, 124, 136, 140, 148, 168, 171, 176, 178–80, 182, 184, 187, 203, 209–11, 219, 232, 234, 257–9 loanable funds, pool of, 1, 39, 58, 124, 136, 182, 187, 203, 209, 219, 232, 234, 257–9 loan demand, 2, 182, 257–9 business, 2, 182, 257 consumer, 2, 257–9 lumpiness, x, 168, 194, 210, 233, 254 M Marginal propensity to consume (MPC), 14, 27, 136, 148, 180, 204, 232 models DSGE, xii, 23, 25–8, 31 structural, 19, 28–31, 36, 90, 121, 124, 125, 132, 139, 220 VAR, 18–25, 31 money supply, M1, 40, 59, 71, 171, 179, 209 money supply, M2, 40, 44 Mountford, A., 20–3 MPS, 137, 232 multicollinearity, xii, 15, 26, 29, 37, 39, 47–50, 52, 56, 73, 98, 111–21, 221, 223, 224, 226, 227, 241–4, 246, 259 multiplier, 6, 8, 12, 26–8, 176, 180, 181, 199, 200, 202, 203, 217, 218, 257 N National Bureau of Economic Research (NBER), 24, 206, 220 national income, 44, 215 NBER See National Bureau of Economic Research (NBER) INDEX 271 Net National Product (NNP), 12–14, 17, 18 Newey West standard errors, 29, 37 nonrecession periods, 206–8, 219–37, 239–49, 252, 259, 187 NYSE composite average, 40, 59 2SLS, Retained profits, 13 Ricardian, 13, 39 robustness, 13, 18, 37, 51, 72, 85–90, 130–2, 182 robustness test, 85, 88 O Obama stimulus, 176, 181, 187, 192, 193, 195, 196, 258 OLS models, 43, 46–62, 91, 93–106 OMB, 19 Ordinary least squares (OLS), 12, 254 other spending, 53, 242 S Sargan test, 6, 15, 29, 64–9, 78–83, 108–10, 112–21, 125, 135, 140–3, 145, 146, 153, 155, 158, 159, 162, 163, 165, 182, 185, 220, 221, 226, 227, 229, 230, 244–6 St Louis equation, 20, 22 scientific, xi, 11, 22, 35, 63, 132, 181 serial correlation, 7, 29, 36, 37, 195, 221 shock, 18, 20, 22, 23, 25, 199, 201, 202 Sims, Christopher, 20, 21, 31 2SLS See Two Stage Least Squares (2SLS) 2SLS models, 37, 63–84, 87, 106–27, 133, 134, 138, 144, 151, 161 Solow, Robert, 132, 195, 199, 203, 204 specification, 13, 14, 17, 26, 35, 37, 47, 89, 98–100, 102, 114, 116 Spencer, R.W., 12 standard error, 13, 15, 17, 29, 37, 38, 183, 184, 186, 236 state and local taxes, 241, 243, 244 state & local revenue, 34, 43 stationarity, 2, 15, 17, 29, 36, 37, 220, 221 stimulus, xi, xii, 1, 2, 5–15, 20, 23, 27, 29, 34, 35, 39, 53, 60, 62, 73, 86, 87, 89, 105, 130, 132, 167–9, 171, 173–97, 200, 201, 204–6, 208–14, 219, 220, 225, 231, 233–7, 251–4, 256–8, 260 stimulus effect, xii, 2, 7–9, 12–15, 27, 34, 35, 39, 53, 87, 132, 173, 174, P Pindyck & Rubinfeld, 36 pool of loanable funds, 1, 39, 58, 124, 136, 182, 187, 203, 209, 219, 232, 234, 257–9 population, 18, 28, 36, 44, 67, 77, 92, 108, 109, 112–21, 124, 192, 204, 211 private spending, 2, 3, 9–11, 15, 24, 29, 39–41, 48, 86, 167, 168, 171, 174, 175, 178, 180, 181, 184, 185, 191, 209, 213, 234, 252, 253, 260 profits, 7, 13, 28, 33, 40, 64, 93, 98–100, 102, 105, 108, 114, 117, 119, 122, 125, 126, 152, 156, 158, 161, 165, 166, 217, 218, 221, 253, 258 R recessionary periods, 2, 222, 244, 258 recessions, 2, 9, 12, 34, 47, 175, 204, 208, 219–37, 239–49, 252 reduced form equations, 19 regression OLS, 12, 15, 46, 63, 122, 127 www.ebook3000.com 272 INDEX 176, 179–81, 183, 185, 188–93, 220, 234–6, 252, 254, 257, 258 stimulus programs, xi, xii1, 11, 60, 73, 86, 89, 167, 168, 171, 175–97, 204–6, 208–14, 219, 225, 231, 233–6, 251–3, 256–8, 260 stimulus programs, deficit financed, 167, 176, 182, 184, 187, 211, 257 strong instrument(s), 34, 36, 37, 67, 68, 71, 72, 75, 77–85, 107, 109–15, 117–21, 125, 135, 136, 140, 141, 143–6, 148, 149, 153, 155, 158–60, 162, 163, 165, 166, 169, 173, 177, 210, 217, 222, 225–7, 229, 230, 244–6, 254, 255 structural, 2, 5, 12–14, 19, 23, 24, 28–32, 35, 36, 63–5, 77, 90, 107, 110, 112, 121, 124, 125, 132, 137, 139, 156, 220 structural equations, 64, 107, 137 supply, 19, 23, 40, 44, 58, 59, 71, 76, 124, 140, 168, 171, 179, 209, 219, 259 supply side, 140 T tax cut, xii, 3, 6, 7, 9, 12–17, 21, 29, 34, 37, 43, 53, 57, 58, 60, 85, 86, 91, 105, 131, 133, 134, 136, 139, 142, 144, 146–9, 151, 152, 166–9, 173–7, 180, 181, 183, 185, 186, 188–93, 195, 196, 202, 210, 218, 220, 223–5, 227, 229–237, 239–42, 244–7, 249, 252, 255–9 tax rate, 12, 13, 18, 23 Time trend, Tobin’s q, 33, 98–102, 104, 105, 108, 114, 119, 152, 158, 211 transfer spending, 15, 16, 53, 55, 56, 176, 191, 239–45, 247, 248, 258, 259 Two Stage Least Squares (2SLS), 6, 28, 29, 35–7, 43, 46, 63, 64, 66, 73, 75–91, 106, 108–10, 112–31, 133–47, 149, 151–6, 158–66, 170, 173, 177, 182, 185, 188, 189, 210, 217, 222, 223, 225–30, 244–6, 248, 254–6 U Uhlig, H., 20–3 unemployment rate, 15–17, 34, 38, 39, 46, 50, 52, 53, 55–7, 60–3, 65, 70–2, 74, 81, 82, 87, 89, 92, 93, 96, 100, 103, 104, 116, 122, 125, 126, 133–7, 142, 143, 151, 154, 156, 163, 176, 177, 189, 191–7, 242, 243, 245, 246, 258 unit root test, 36 utility function, 25 W Wald test, 6, 64, 65, 67, 77, 78, 80–2, 107–10, 112–21, 125, 135, 153, 155, 159, 182, 220, 222, 229, 230, 244 weak instruments, 6, 34, 36, 37, 44, 64, 67, 68, 70–2, 77–81, 83, 107–10, 112–16, 119–21, 135, 142, 144, 146–9, 182, 220, 244 wealth, 3, 13, 28, 33, 44, 45, 47, 52, 62, 75, 92, 93, 140, 178, 201, 221, 225 Woodford, M., 27 Y Yohe, W.P., 12 young/old ratio, 36, 44, 67, 77, 108, 211 .. .Crowding Out Fiscal Stimulus John J Heim Crowding Out Fiscal Stimulus Testing the Effectiveness of US Government Stimulus Programs www.ebook3000.com John J... show what they can In theory, the effectiveness of stimulus programs can be curtailed by “crowd out Crowd out theory suggests that whatever their stimulus effects, government deficits have the undesirable... stimulates the economy, according to stimulus theory In testing whether these stimulus programs actually work, we took care to test equations taken from the Keynesian model This gives the stimulus effects