Investment and Profits PDF
Preview Investment and Profits
Investment and Profits: Causality Analysis in Selected eu Countries IgorStubelj UniversityofPrimorska,Slovenia [email protected] Thepurposeofthepaperistopresentthetheoreticalandempiricalanal- ysisofthecausalrelationshipbetweencompany’sinvestmentandprofits. Followingthepreviousresearchfindingsthatarenotunanimous,weex- aminedthecausalrelationbetweenprofitsandinvestmentbothattheag- gregateandattheindustrylevel.Thetheoryonthefieldallowsustodefine threebasictheses:profitsdetermineinvestmentontheaggregatelevelor industrysectorslevelinanationaleconomy,investmentdeterminesprof- its,andprofitsdetermineinvestmentandinvestmentdetermineprofits.We usedtheGrangerdefinitionofcausalityandtheToda-Yamamotoproce- dure.Basedontheresultsoftheanalysis,wecanclaimthatinvestmentde- terminesprofitsinanationaleconomy.Attheindustrysectorslevelthere isthegreatestsupportforacause-and-effectrelationinthedirectionfrom profitstoinvestment.Theresultsofourempiricalanalysisandcontextual interpretation can help in identifying potential factors and barriers that hindertheeffectivefunctioningoftheeconomicsystem.Itisundisputed thatinvestmentandprofitsisextraordinarilyimportantfortheeconomic systemandthattheinvestmentandprofitsrelationisundeniablystrong. KeyWords:investment,profits,causality,industry jel Classification: e22, g31 IntroductionandTheoreticalBasis Classical economists as Smith (1805) and Ricardo (1817) emphasize the importanceofinvestmentineconomicgrowth.Keynes(1936)builtanew paradigmoninvestment,connectedwithothereconomiccategories.For Keynes investment is a prevailing factor and an important determinant ofnationalproduct. For the Keynesianeconomistsinvestment depends onprofitexpectationsthat,besidedemandandinstitutionalfactors,are based upon ‘animal spirits’ (Stockhammer 2006). Asimakopulos (1971) explained the Keynes investment model. He presents his point of view, showingatwo-waydirectionofcausalitybetweeninvestmentandprofits. Profitexpectationsplayacrucialroleinfirms’investmentdecisionsand ManagingGlobalTransitions12(4):395–413 396 IgorStubelj theseexpectationsarestronglyaffectedbycurrentinvestmentlevelsince thisisanimportantdeterminantofcurrentprofits. Jorgensonandhiscollaborators(HallandJorgenson1967)contributed to neoclassical theory with the development of the neoclassical invest- mentmodel.Theneoclassicalinvestmentmodelstartingpointistheso- lutionofdynamicproblemthatdeterminesthecompanies’capitalneeds trough time. Neoclassical theory somehow shows the causal link from profitstoinvestment(Gupta1988). For Post-Keynesians, investment is a very important determinant of theeconomicsystem.Theargumentisthattheexploitationofnewtech- nologiesispossibleonlythroughinvestment.WhatKalecki(1942;1971) wantedtopointoutwasthatthemostimportantprerequisitetobecome anentrepreneuristhecapital.Withthisstatement,Kaleckisuggeststhat the investment determines profits. The focus is on companies’ physical capitalinvestment.TheimportancethatPost-Keynesiansassignedtoin- vestmentraisesakeyquestionthatreadsasfollows:whatdeterminesthe investment level? Joan Robinson (1962) has developed a model starting fromKaldor’s(1955)modelwithaninvestmentwithinthemodel.Compa- nieschooseinvestmentsbasedonexpectedprofitmargins.Theexpected profitmarginsexplaintoalargeextenttheactualprofitmargins.Robin- son set a two-way relationship between investment and profits with his formula.Hertheorystatesthatinvestmentsdependonprofitsandprofits on investments. Post-Keynesian research in the field of investment has continuedontheimportanceofinternalfinancing.Thiswasincontrast withModiglianiandMiller(1958)theoremthatundercertainconditions emphasizedtheirrelevanceoffinancingpolicyandfinancialstructurefor thecompanies’investmentandmadeatheoreticalframeworkforthere- search of the importance of financial factors for the companies’ invest- ment(Stockhammer2006). EmpiricalEvidence Notwithstanding the key role of companies’ sustainable investment for the expected cash flows and companies’ value, sectors and the whole economy, just a few important researches examined the causal link be- tweeninvestmentandprofitsandshowedthattheinvestmentisvitalfor theexpectedprofits.Theresultsofempiricalstudiesareunclear;fromthe researchresults,itisdifficulttounequivocallyconcludethedirectionof a causal link between profits and investment or cash flows. Studies car- riedoutbyBaumoletal.(1970),Little(1962),FriendandHusić(1973)do not support the hypothesis that profits results from past investment. In ManagingGlobalTransitions InvestmentandProfits 397 theirstudiesBar-Yosefetal.(1987)foundthatinvestmentdoesnotcreate profits, but profits raise investment, defined in terms of Granger causal effectrelationship.Theyalsofoundthattheprofitsareadeterminantof companies’ business investment. Shapiro, Sims, and Hughes (1983) and McFetridge(1978)surveyssupportthehypothesisthatcashflowsresult frompastinvestment.IntheanalysisofMahdavi,Sohrabian,andKholdy (1994) about the causal effect relationship, employing the ecm (Error CorrectionModel),whichtakesintoaccountthecointegrationbetween realcashflowsandrealinvestmentofcompanies,theyfoundthatthere is evidence of the one-way causal effect relationship of real companies investment to real cash flows. This is in contrast to some of the previ- ous studies that demonstrated that the profit or cash flows at the level ofcompaniesorindustriesdetermineinvestment.LeeandNohel(1997) haveconcludedthatthereisatwo-waylinkbetweeninvestmentandprof- its. Gupta (1988), who based on two different methodologies, comes to different conclusions about the direction of causality. Akyüz and Gore (1996),andInci,Lee,andSuh(2009)haveconcludedthatthereisalink inthedirectionearningstoinvestment. Asseen,theempiricalstudiesdonotgiveasingleanswertotheprob- lemaboutthedirectionofthecausallinkbetweeninvestmentandprof- its. The reasons for the different findings may lie in a variety of used methodological approaches, information restrictions and limitations of the studies. Previous researches are based on different samples of data, both in terms of time-scale and different economic backgrounds (most us based). Different economic and regulatory environment can play a crucialinfluenceonthedecisionmakingofmanagersandthusthelink between investment and profits. Different economic environments may involveavarietyofconditionsthatarenecessaryforefficientinvestment, whichofcourseaffectstheambitiousproblemofcausality. Onthebasisofpreviousresearchwecanconcludethatthefindingsof studiesinwhichdatawereusedattheaggregatelevel,forthemostpart support the causal effect relationship in the direction of investment to profitsortwo-waylink.Researchbasedondataatthefirmlevelislargely supported by causal effect relationship in the direction of the profits to investmentsortwo-waylink. ThesesandHypothesesforEmpiricalTesting If we realize a synthesis of theoretical concepts framed in micro-and macroeconomictheory, thetheoryof financeandentrepreneurship,we mayjustifyacorrelationlinkbetweeninvestmentandprofits.Thisthe- Volume12·Number4·Winter2014 398 IgorStubelj oretical framework also allows us to define the causal link between the exposed variables, which runs from profits to investment, from invest- menttoprofitsorinbothdirections. Thethreebasicthesesare:profitsdetermineinvestmentontheaggre- gateandindustrysectorslevelinanationaleconomy,investmentdeter- mines profits, and profits determine investment and investment deter- mine profits. The thesis of the causal link, running from profits to in- vestment,hasalimitedexplanatorypowerbecauseitisonlyvalidinthe special case when the primary companies’ objective in the economy is to maximize profits. This means that companies should only invest in thecurrentandexpectedmostprofitableactivities.However,ifthemaxi- mizationassumptionisnottrue,theninvestmentisinducedalsobyother motives,whichareindependentoftheprofitmotive.Thedoubtoftheva- lidity of assumptions about theprofitmotive is offeredparadoxicallyin the economic theory. In particular the part of economic theory which refers to the business sciences, clearly explains that the company also followsothergoals,suchasrevenuemaximization,maximizationofthe market share, customer satisfaction, production optimization ... (Janeš andDolinšek2010;NovakandŽižmond2012;Odar,Kavcic,andJerman 2012;Radosavljevićetal.2011).Throughtheseobjectives,managersand equity holders can achieve the purpose of the company, which is to in- creasethevalueofacompany’sequityandthusshareholdervalue.Crotty (1992;1996)wrotethatcompaniesdevelopandadoptrulestohelpthemto copewithuncertainty.Gordon(1992)arguedthattheprimaryobjective of companies is a long-term survival. Many authors from the last Post- Keynesianperiodarguethatthegoalofmoderncorporationsisnotprofit maximizationbutalong-termsurvivalandgrowth(Stockhammer2006). Thismeansthatinvestmentisprimarilynotdependentonprofits.On contrary, investment determines profits. These arguments allow us to claimthatprofitsdependoninvestmentandnotviceversa.Byconfirm- ingthethesisthatprofitsresultfrominvestmentontheleveloftheentire economy,wemayimplythatprofitabilitydoesnotdetermineandallocate investmentbuttheallocationofinvestmentdeterminesprofitability. Imperfectionsoffinancialmarketsthatincreasethedifferencebetween internal and external financial resources act in contrary to the theory- basedinterpretation,whichdefinesthedirectionofcausalityfromprof- itstoinvestment.This meansthatthefailureoffinancialmarketsaffect theallocationofcapital.Ontheonehand,thelimitedfinancialresources andthedisparitybetweentheexternalandinternalsourcesofcapitalaf- ManagingGlobalTransitions InvestmentandProfits 399 fectthelevelofinvestmentandworkinfavouroftheprofitmotiveofin- vestment.Companiesmustachievehigherprofitability,investinginmore risky and profitable activities. On the other hand transaction costs, the informationasymmetryproblemandtheriskalllimitsthemigrationof capitalandgocontrarytothegainfulmotive.Companiesinvestthecre- atedin-house resourcesinthemwellknown availableandlessrisky in- vestmentsatthecostoflowerprofitability.Inthisway,companiespursue agoaloflong-termsurvival.Giventheseimplicationsandtheresultsof research,itisdifficulttodefinetheimpactthattheincompletenessoffi- nancial markets have on the direction of causality between investment andprofits. In order toconfirm the theses we set the following hypotheses at the aggregateandindustrysectorlevelinanationaleconomy: h1 Profitsdetermineinvestment. h2 Investmentdeterminesprofits. h3 Investmentdeterminesprofitsandprofitsdetermineinvestment. DataandMethodology WegatheredthedataforinvestmentandprofitsfromtheEuropeanCom- mission,Eurostatstatisticdatabase.Fortherepresentationofinvestment (I)weusetheGrossfixedcapitalformation(esa95,3.102)whichaccord- ingtoEurostatdefinitionconsistsofresidentproducers’acquisitions,less disposals,offixedassetsduringagivenperiodpluscertainadditionsto the value of non-produced assets realized by the productive activity of produceror institutional units. For therepresentationof profits(P), we useNetoperatingsurplusandnetmixedincome.Netoperatingsurplus isinterpretedasthereturntocapitalortheeffectoftimepassingonthe netpresentvalue.Forunincorporatedenterprises,ownedbyhouseholds, thiscomponentiscalled‘mixedincome.’ WewereabletoobtainthedataonthenationaleconomylevelforAus- tria,Denmark,Italy,FinlandandNetherlands.Wegatheredtheindustry leveldataforAustria,Italy,FinlandandNetherlands.Forthesecountries, itispossibletogetalongtimeseriesofdatafrom1977to2007toperform thetestsofcausality.Weusedthe nace31industryclassification.Weend upwith 88 time series (22 industries for eachcountry) foreachof both variables. Data for some industries was not available and we left out of our analysis aggregates of industry groups. To obtain data in real terms wedeflatedthevariablenetoperatingsurplusandnetmixedincome.All Volume12·Number4·Winter2014 400 IgorStubelj thedataareinvaluelevelsinMillionsofeuro,chain-linkedvolumes,ref- erenceyear2000(at2000exchangerates). We can define the causal relationship between two variables, invest- mentandprofitsinourcase,withtheGrangerdefinitionofcausality.It shouldbenotedthattheassertionthatXcausesYinGrangersensedoes notimplythatthevariableYistheresultofthevariableX.Grangercausal- itymeansanteriorityofdatafluctuationsanddoesnotimplycausalityin thegeneralsenseofthewordandintermsofcontent. The implementation of Granger causality tests is connected with the problems of stationarity and cointegration properties of the considered timeseriesofdata.Grangercausalitytestsaresensitivetonon-stationary timeseries.Thefirststepintheanalysisofcausalityisthecheckforsta- tionarityandtheintegrationdegreeofvariables.Thesecondstepisiden- tifyingthepossiblecointegrationofvariables(Bekő2003). Only in the third step, we apply the analysis of causality. Standard Granger tests of causal effect relationship are valid only if the original timeseriesarenotcointegratedandmustbeappliedonstationaryvari- ables. When this is not the case, we must differentiate the variables to reachstationarityand,incaseofcointegration,wemustperformthetest witherrorcorrectionmodels. We can apply the test developed by Toda and Yamamoto (1995) and avoid the procedure of testing for cointegration. Testing for cointegra- tionandapossibleneedfordifferentiationtoreachstationaritycouldbe problematic, as we lose some information with the differentiation. The inferenceofcointegrationcouldbeunreliable.Bothproblemsmayaffect theconclusionsofthecausalitytests.TheTodaandYamamotoprocedure consistsintestingthevectorautoregression– var systemequations: (cid:8)m m(cid:8)+el (cid:8)m m(cid:8)+el Pt = aiPt−1+ bjPt−j+ ciIt−i+ djIt−j+εt (1) i=1 j=m+1 i=1 j=m+1 (cid:8)m m(cid:8)+el (cid:8)m m(cid:8)+el It = eiIt−1+ fjIt−j+ giPt−i+ hjPt−j+ηt, (2) i=1 j=m+1 i=1 j=m+1 where It is investment for period t represented with Gross fixed capital formation,Pt isprofitsforperiodt,representedwithnetoperatingsur- plusandnetmixedincome,lettersfromatohareindependentvariables coefficients, m is period lags, εt and ηt are regression errors, el are ex- traperiodlags.Toapplythetestwemustdefinethenumberoflagsand ManagingGlobalTransitions InvestmentandProfits 401 table1 TheResultsoftheCausalityAnalysisattheNationalLevelforAustria, Denmark,Finland,ItalyandNetherlands Country P→I Significance I→P Significance Austria No . No . Denmark No . Yes . Finland Yes . Yes . Italy Yes . Yes . Netherlands No . Yes . the number of extra lags. I Granger cause P if any of the ci coefficients isstatisticallydifferentfrom0.PGrangercausesI ifanyofthegi coeffi- cientsisstatistically differentfrom0.For thetestofthenullhypothesis that every ci = 0 or every gi = 0 we perform a Wald statistic test. To avoidtheproblemofarbitrarydecidedlagswedefinetheoptimalnum- beroflagswiththefollowingtestsfor var:sequentialmodified lr test statistic,finalpredictionerror,Akaikeinformationcriterion,Schwarzin- formationcriterion,Hannan-Quinninformationcriterion.Theproblem ofsuchtestsisthatwemustdeterminethenumberofmaximumlagsfor testing which influence the results. We select the optimum lags in base of the most frequentoptimal lag, resulting from all the tests performed from1upto6lags.WeusetheAugmentedDickey-Fullerunitroottest equationtodefinetheorderofintegrationofthevariables.Theextralags equaltotheorderofintegration. ResultsandDiscussion Wepresenttheresultsoftheanalysisofcausalityintables1,2and3.We examinedthehypothesesattheaggregatelevel.Basedontheresults,we cannotconfirmthefirsthypothesiswiththecriterionofmorethan50. We found the causal link from profits to investment just in two of five countries. The causal link from investment to profits is stronger as we found it in four of five analyzed countries. We can confirm the second hypothesisthatinvestmentdeterminesprofitsattheaggregatelevelina national economy. We cannot confirm the third hypothesis that invest- ment determinesprofitsand profitsdetermine investment at the aggre- gatelevelinanationaleconomy.Basedontheresultsoftheanalysison theaggregatelevelofanationaleconomy,wecansaythatinvestmentde- terminesprofitsinanationaleconomy. Theresultsattheaggregatelevelareinlinewiththeempiricalfindings Volume12·Number4·Winter2014 402 IgorStubelj in the researches carried out by Gupta (1988) and Mahdavi, Sohrabian, and Kholdy (1994). The results are also in line with the findings of the researches done by Akyüz and Gore (1996), Lee and Nohel (1997) and Heshmatiand Lööf (2006) whereauthors found out the causal link be- tweeninvestmentandprofitsinbothdirectionsinapartwhichshowsthe causallinkfrominvestmenttoprofits. Basedontheresults,weexaminedthehypothesesattheindustrysec- torlevel.In28ofindustries,wefoundacausallinkfromprofitstoin- vestment.Basedonthecriterionofmorethan50industries,inwhich causalityrunsfromprofitstoinvestment,wecannotconfirmthefirsthy- pothesis,sayingthatprofitsinindustrysectorsdetermineinvestmentin industrysectors.In14ofindustries,wefoundthecausallinkfromin- vestmenttoprofits.Basedonthecriterionofmorethan50industries, inwhichcausalityrunsfrominvestmenttoprofits,wecannotconfirmthe secondhypothesis,sayingthatinvestmentsinindustrysectorsdetermine profits in industry sectors. In 19 of industries, we found a causal link, runninginbothdirections.Basedonthecriterionofmorethan50in- dustries, in which causality runs in both directions, we cannot confirm the third hypothesis, saying that investment in industry sectors deter- mineprofitsinindustrysectorsandprofitsinindustrysectorsdetermine investmentinindustrysectors. However, the results give stronger support to a causal relation from profitstoinvestment.Thisisconsistentwiththefindingsoftheresearch carriedoutbyBar-Yosefetal.(1987)andIncietal.(2009),inwhichthe authorsfoundthatprofitscauseinvestmentinGrangersense,orthatthe causal link in the direction from investment to profits is weak. Our re- searchfindingsareinlinewithempiricalresearch,carriedoutbyAkyüz and Gore (1996), Lee and Nohel (1997) and Heshmati and Lööf (2006), inwhichtheauthorsfoundatwo-waycausallinkbetweentheresulting investments and profits in the part for the causal relationship, resulting inthedirectionfromprofitstoinvestment. The results of our empirical analysis gave us just the causal link be- tween the resulting investment and profit in a Granger sense. This did notmeanthatinvestmentsleadtoprofits,profitsleadtoinvestment,or bothinacommonsenseorinasenseofcontent.Grangersensecausality doesnotimplythattheinvestmentorprofitsresultfromprofitsorinvest- ment,orboth.Grangercausalitysimplymeasurestheanteriorityofdata fluctuations.Itisanunstructuredmodelofevidencethathasnosubstan- tiveinterpretation.Thealternativeisastructuredmodelthatexplainsthe ManagingGlobalTransitions InvestmentandProfits 403 table2 TheResultsoftheAnalysisofCausalityinIndustriesforAustria,Finland, ItalyandNetherlands Industry a fi it nl Agriculture,huntingandforestry P↔I P↔I I(cid:2)P I→P Miningandquarrying P↔I I(cid:2)P I(cid:2)P I→P Manufactureoffoodproducts;beveragesand I(cid:2)P P→I I(cid:2)P I→P tobacco Manufactureoftextilesandtextileproducts I(cid:2)P P→I I(cid:2)P I(cid:2)P Manufactureofpulp,paperandpaperproducts; P→I P↔I I→P P→I publishingandprinting Manufactureofcoke,refinedpetroleumproducts P→I P→I I(cid:2)P I(cid:2)P andnuclearfuel Manufactureofchemicals,chemicalproductsand P↔I P→I P→I I(cid:2)P man-madefibres Manufactureofrubberandplasticproducts I(cid:2)P P↔I P↔I I(cid:2)P Manufactureofbasicmetalsandfabricatedmetal I(cid:2)P P→I I(cid:2)P I→P products Manufactureofmachineryandequipment P→I P↔I I(cid:2)P P→I Manufactureofelectricalandopticalequipment P↔I P↔I I→P I(cid:2)P Manufactureoftransportequipment I(cid:2)P I(cid:2)P I→P I→P Manufacturing I(cid:2)P P↔I I(cid:2)P P→I Electricity,gasandwatersupply P→I I(cid:2)P P→I I→P Construction I(cid:2)P I→P I(cid:2)P I(cid:2)P Wholesaleandretailtrade;repairofmotorvehicles, I(cid:2)P P↔I I(cid:2)P P→I motorcyclesandpersonalandhouseholdgoods Hotelsandrestaurants I(cid:2)P P↔I P↔I I(cid:2)P Transport,storageandcommunication P↔I P→I P→I P↔I Financialintermediation P→I P→I P→I I(cid:2)P Realestate,rentingandbusinessactivities P→I I(cid:2)P I(cid:2)P P→I Healthandsocialwork I(cid:2)P P→I I→P P→I Othercommunity,social,personalserviceactivities I(cid:2)P P↔I P→I I→P notes a–Austria,fi –Finland, it –Italy, nl –Netherlands.P ↔ I –twosided causalitybetweeninvestmentandprofits,P → I–causalityrunningfromprofitstoin- vestment,I→P–causalityrunningfrominvestmenttoprofits,I(cid:2)P–nocausality. mechanismsandfactorsthroughwhichonevariableinfluencesanother. Theadvantageofthestructuredmodelisthatwemakeaconceptualex- planationofaphenomenon.Thereisariskthatthemodelisnotcorrectly Volume12·Number4·Winter2014 404 IgorStubelj table3 AssembledResultsofCausalitybetweenInvestmentandProfitsfor IndustriesintheAnalyzedCountries Country P↔I P→I I→P I(cid:2)P Austria Finland Italy Netherlands Share . . . . specifiedandleavesoutimportantrelevantfactors.Themethodologyal- lows us to analyze Granger causality through the unstructured model. Theadvantageofunstructuredmodelisthatitdoesnotplacerestrictions onhowandwhyonevariableaffectsanother.Thedisadvantageisthatthe relationshipbetweenvariablesdoesnothaveanycontentvalue.Thus,it remainsachallengingtasktorightlyinterprettheanalysisresultsofthe causal link in Granger sense. We linked the results of the unstructured modelwiththetheoreticalbasisandtheresultsofpreviousempiricalre- searchonthisfield.Wedescribethelogicbehindtheanalyzedresultsand theimplicationsoftheresultsfortheinvestmentpolicy. Theresultsthatinvestmentdeterminesprofitsatthenationallevelhave several implications. From the financial theory point of view we could assumethattheanalyzedcountrieshaveawell-developedfinancialsys- temandtheinnovationsoffinancialinstrumentsdiminishtheproblem ofobtainingexternalfinancialresources.Financingproblems,described inworksofKeynes(1936),Kalecki(1971),andMinsky(1975),donotplay acrucialrole.Eveninthecaseoffinancialconstraintsofcompanies,other companiesthatarenotfinanciallylimitedcantakeoninterestinginvest- ment,andthusgenerateprofitsthatsupportthecausalrelationshipinthe direction of investment to profits. For the support of such thinking see FurlongandWeiss(1990).Transactioncostsarenotanimportantlimita- tiontocapitalreallocation. Mukherjee and Henderson (1987) have considered whether, in prac- tice,theexpectedandrequiredreturnsaretruekeyfactorsofinvestment decisionofcompanies,orsomeotherfactorsaremoreimportant.They foundthatinpracticemanyprojectsarerejectedforreasonsotherthan economicones.Guidelinesanddecisionrulesshouldleadtoinvestment decisionsthatenabletofulfiltheprimaryobjectiveofthecompany,that is, to increase the value of a company’s equity. Despite this, in practice ManagingGlobalTransitions
Description:The list of books you might like
