Business Investment in the 1970S: a Comparison of Models

Business Investment in the 1970S: a Comparison of Models

CHARLES W. BISCHOFF Yale University Business Investment in the 1970s: A Comparison of Models THE INVESTMENT BOOM THAT PERSISTED throughout the late nineteen- sixties is now over. Whether private spending for fixed capital, after allow- ance for rising prices, will increase at all in 1971 is an open question at this point. This paper attempts to supply an answer and to look a bit further ahead, into 1972 and early 1973. In order to do this, a variety of different models is presented and pro- jected into the future. This is necessary because to date no consensus has de- veloped among economists about the determinants of investment (here taken as private expenditures on nonresidential fixed capital-plant and equipment) or about the magnitude and timing of the effects of mone- tary and fiscal policies on this aggregate. As Arthur Okun has pointed out: The best example I can offer [of a purely scientific,nonideological controversy among economists]is the disagreementamong studentsof businessinvestment regardingthe relative importance of internal cash flow, the cost of external capital, and the growth of final demandas determiningfactors.' This paper also extends the preliminary attempt to explain the behavior of investment spending in 1969-70 that I reported on a year ago.2 Many models of investment behavior have been advanced, but only a few 1. Arthur M. Okun, The Political Economy of Prosperity (Brookings Institution, 1970), p. 19. 2. CharlesW. Bischoff, "Plant and EquipmentSpending in 1969 and 1970," Brook- ings Paperson EconomicActivity (1:1970), pp. 127-33. 13 14 Brookings Papers on Economic Activity, 1:1971 researchershave attemptedto comparethem systematically.3The workre- portedhere is confinedto a veryhigh degreeof aggregation,which severely limits the possibilityof sharpdiscrimination between models. I can hope only to highlightthe range of disagreementand perhapsmore important, to suggestthe extentto whichit leads to differentpolicy prescriptions. This paper presentsempirical characterizations of a numberof points of view andreports (a) how well they explainthe investmentexperience of the post- Koreanera, with particularemphasis on 1969 and 1970;(b) the extent to whichthe modelsdiffer with respect to the way monetaryand fiscal policies affectinvestment, and the extentto whichthese differences are essential and economicallyimportant; and (c) whatthe modelshave to say aboutcapital spendingprospects over the next two or threeyears, given severalalterna- tive scenariosfor the path of the economyas a whole. The Models Five differentsets of equationsexplaining investment behavior are pre- sentedand applied in this paper.This selectionis representativeof fivetheo- reticalpositions on the demandfor fixedcapital goods. In each case I have separatedinvestment in equipmentfrom investment in nonresidentialstruc- tures,primarily because the tax policiesapplied to these two types of asset have divergedgreatly in the past few years.Other factors differentially af- fectingplant spendingand equipmentspending, such as the pricedeflators for the respectiveaggregates, have also moved very differently.4Except in one case, however,the form of the equationis the samefor both plant and equipment. 3. Prominentstudies involving systematiccomparisons include Dale W. Jorgenson, JeraldHunter, and M. Ishag Nadiri, "A Comparisonof AlternativeEconometric Models of QuarterlyInvestment Behavior," Econometrica, Vol. 38 (March 1970), pp. 187-212; Jorgensonand Calvin D. Siebert,"A Comparisonof AlternativeTheories of Corporate InvestmentBehavior," American Economic Review, Vol. 58 (September1968), pp, 681- 712; and Edwin Kuh, Capital Stock Growth:A Micro-econometricApproach (Amster- dam: North-Holland, 1963). 4. The price deflators for any given quarter are taken as predetermined,implying that, at least for the currentperiod and for the rangeof demandslikely to be encountered, supply is infinitelyelastic at the given price. With this proviso, all of the equations can be treated as proper demand equations. In the simulations of the future, these prices are based on price forecasts from a complete econometric model, in which the prices dependprimarily on unit labor costs or wages and on price indexesfor the economy as a whole, but also partially on the share of investmentin total output, with a distributed lag starting two quartersback. CharlesW. Bischoff 15 For threeof the equations,two from one model and one from another,I presentan alternativeversion, which incorporates capital gains, a factornot includedin the standardequations. The three additionalequations bring the total to thirteen.Although I mentionvarious economists in discussing the rationalebehind each of the equations,I must emphasizethat the pre- cise formalspecification of the equationsis solely my responsibility. THE GENERALIZED ACCELERATOR MODEL The most venerablemodel, with antecedentsgoing back at least to J. M. Clark,5is basedgenerally on the accelerationprinciple, which postulates a linearrelationship between net investmentand changes in output.As modi- fied and generalizedby, amongothers, Chenery, Koyck, Eisner,and Hick- man, the model has developedas a generaldistributed lag relationship involvingboth changesin and level of output, along with the level of the existingcapacity or capital stock.6In one of a numberof formulations fallingwithin this class,a firmforms expectations about its futureoutput on the basisof the past output(or sales)of the firmitself, the industryto which it belongs,or both. It then makes plans to adjustits capitalstock toward the level that would be an optimumfor producingthe plannedoutput, if this outputwere to representa long-runequilibrium. If the adjustmentin any given periodis not complete,this can be rationalizedin terms of the additionalcosts that the firmwould incur if it tried to make a very rapid adjustment,and in terms of uncertaintythat futuredemand will prove to justify the plans.7 5. J. Maurice Clark, "BusinessAcceleration and the Law of Demand: A Technical Factor in Economic Cycles," Journalof Political Economy,Vol. 25 (March 1917), pp. 217-35. The antecedentsgo back even earlier.See C. F. Bickerdike,"A Non-Monetary Cause of Fluctuationsin Employment,"Economic Journal, Vol. 24 (September1914), pp. 357-70. 6. Hollis B. Chenery,"Overcapacity and the AccelerationPrinciple," Econometrica, Vol. 20 (January1952), pp. 1-28; L. M. Koyck, DistributedLags andInvestment Analysis (Amsterdam:North-Holland, 1954); Eisner's work on this subject,going back to 1952,is extended and summarizedin Robert Eisner, "A PermanentIncome Theory for Invest- ment: Some EmpiricalExplorations," American Economic Review, Vol. 57 (June 1967), pp. 363-90; Bert G. Hickman, IMvestment Demand and U.S. Economic Growthl(Brook- ings Institution, 1965). 7. The formalconsideration of these adjustmentcosts, as well as the costs of adjusting other factors of production, leads to extremelycomplicated theoreticalformulations, and to equations that look quite differentfrom the one I am presenting.The addition of the problemof uncertaintymakes the theoreticalformulations still more formidable. All of the equations in this paper are intended to approximatemodels of investment behaviorthat have actuallybeen applied,and thus I have not consideredthe theoretically 16 BrookingsPapers on EconomicActivity, 1:1971 The idea of a partialadjustment process is commonto all of the models presented,and will not be repeatedas each model is introduced.The dis- tinguishingfeature of the acceleratormodel is that the determinationof the plannedcapital stock is based only on output, and not on such factorsas the cost of capital, the price of investmentgoods relativeto wages, and variousfeatures of the tax system.This pure dependenceon output may resultfrom technological rigidities that permit only one capital-outputratio for each product.On the otherhand, the modelmay performwell relative to other models, not becauseof such technologicalrigidities, but because the othermodels are deficientin specifyingthe preciseway in whichother factorsdetermine the optimumcapital-output ratio. The model is completedby the assumptionthat replacementinvestment is proportionalto existingcapital stock, or plannedoutput, or some func- tion of the two. The demandsof the individualfirms are summed to obtain demandat the industryor economy-widelevel. Differencesin the capital- output ratio among firms or industrieswill introducethe possibilityof aggregationerror. Althoughfew economistswould consider them complete representations of the investmentprocess, models of this sort have been tested againsta greatervariety of data than have any of the othersunder discussion, and they have generallyperformed well. Many forecastersuse this model as at least one elementin theirpredictive equations, but theyusually modify it by addingother variables,including interest rates, cash flows, and variables designedto incorporatethe effectsof tax policies. The mathematicalstatement of the generalizedaccelerator model that I shall use is shown as equations(1) and (2) in Table 1. THE CASH FLOW MODEL A varietyof theoreticalconsiderations have been presentedfor focusing on profitsor cash flow as a determinantof businessinvestment. Current and past profitsmay be thoughtof as a good proxyfor futureprofit expec- tations, which in turn determineinvestment.8 Given the changes in tax superior but practicallycumbersome models currentlyunder development.The most cogent discussionof the importance,as well as the difficulty,of theoreticalresearch in this area is in

View Full Text

Details

  • File Type
    pdf
  • Upload Time
    -
  • Content Languages
    English
  • Upload User
    Anonymous/Not logged-in
  • File Pages
    51 Page
  • File Size
    -

Download

Channel Download Status
Express Download Enable

Copyright

We respect the copyrights and intellectual property rights of all users. All uploaded documents are either original works of the uploader or authorized works of the rightful owners.

  • Not to be reproduced or distributed without explicit permission.
  • Not used for commercial purposes outside of approved use cases.
  • Not used to infringe on the rights of the original creators.
  • If you believe any content infringes your copyright, please contact us immediately.

Support

For help with questions, suggestions, or problems, please contact us