Credit rationing of Polish farm households. A theoretical and empirical analysis.

On the verge of EU accession, Poland’s agricultural sector is characterised by anumber of distinct structural weaknesses, which are a major reason for the unsatisfactoryincome situation of rural households. Among these weaknesses are thatfarm productivity is substantially below EU standards, investment has performedmuch weaker as compared with the overall Polish economy, and structuralchange has been very small. It has been suggested that credit access is acrucial factor for overcoming these undesired outcomes. Furthermore, the Polishgovernment massively intervenes on rural credit markets, in particular by grantingsubsidies on working capital and investment loans for agriculture. Againstthis background, the aim of the present research is to discover how far potentialdeficiencies on rural credit markets can be made responsible for the structuralweaknesses of the Polish farm sector and thereby provide an economic rationalefor government activity.Central to the analysis is the notion of credit rationing. This notion abounds inthe recent literature on credit market problems, although it is not used in a uniformway. In this monograph, credit rationing is understood to indicate a situationof persistent private excess demand for credit. The subsequent theoreticaland empirical analysis explores how far this concept can be made fruitful for theunderstanding of the Polish rural credit market and the effects of governmentalintervention.The major findings of the study can be summarised as follows:1. The theoretical investigation of credit markets sets out that credit rationing isby definition excluded in the traditional neoclassical market model, whereasit is a likely outcome on markets with asymmetric information. However,theory does not provide unambiguous propositions regarding the welfare assessmentof credit rationing. Since the presence of asymmetric informationcancels the functioning of the price mechanism, the traditional concepts ofsocial efficiency are no longer valid. As a consequence, credit rationing doesnot necessarily imply underinvestment, and it generally does not create a casefor straightforward government policy. It seems therefore reasonable to analyticallydecouple the analysis of credit rationing and under- or overinvestment.The subsequent analysis in this monograph focused on the first ofthese.2. The extent to which asymmetric information has harmful effects on investmentoutcomes is shown to depend on the availability of counteracting arrangements, such as collateral, joint liability, or reputation of borrowers. Theway in which governments can improve on these instruments will play a decisiverole for successful policy action. Any intervention measures shouldconsider the conditions and causes that are responsible for an undesirablemarket outcome, in case that this has been successfully identified.3. In the framework of a two-period farm household model, the consequencesof introducing a binding credit constraint are examined. The market interestrate loses its relevance for the household internal allocation of funds and isreplaced by an endogenous, unobservable shadow interest rate. Comparedwith a first-best world without credit rationing, the household will reduceoutput, which implies a loss of income. An increase in government transfersrelaxes the liquidity constraint and thus has positive effects on farm output.In a multi-period household model with endogenous equity formation, creditrationing has the effect that investment cannot immediately attain its optimallevel. The household thus reduces current consumption in favour of equityformation.4. The presence of a perfect capital market allows the convenient separation ofproduction or investment decisions on the one hand and consumption decisionson the other. Both farm household models suggest that this cannot bemaintained under a binding credit constraint. As a consequence, there is noobjective criterion anymore which allows to assess the (private) efficiency ofinput use or investment activities. Both decision complexes can only be madesimultaneously with the household’s consumption plan and are thus affectedby the household’s preferences. Any empirical production or investmentanalysis has to take these interdependencies into account.5. In a reflection on economic methodology, it is argued that econometrics cannotbe the fundamental benchmark for the falsification of theories. The methodologicalstandpoint of critical rationalism should therefore be left behindand be replaced by a more pragmatic and instrumentalist position. The empiricalresults of the present study are primarily based on a regression analysisof cross-sectional survey data, which includes qualitative and quantitativeindicators of credit rationing.6. According to statements of farmers made during the survey, 80 percent offarm households took at least one loan in the reporting period 1997-1999.Almost half of the borrowers obtained less credit than desired and are henceregarded as credit-rationed. Central determinants of credit rationing are thereputation of the loan applicant as well as demographic household characteristics. Over all loan types, respondents with a good credit history have a 30percentage points lower probability of being rationed than borrowers who rescheduleda loan in the past. In addition, more adult males in the householddecrease the probability of being credit-rationed, while more females increaseit. If only short-term borrowing is considered, collateral availability isan additional key factor of credit rationing.7. The econometric analysis of output supply supports the earlier finding thatmore than 40 percent of borrowers experienced pronounced credit rationingby rural banks. These farms display a marginal willingness to pay for creditof on average 209 percent net of principal. The willingness to pay is significantlydifferent from individual interest rates for credit that account for loanspecific transaction costs. These individual interest rates are 13 percent perannum on average. Transaction costs, however, do not ex-post rationalise awithdrawal of loan applications and cannot be regarded as the ultimate causeof perceived credit rationing. In the group of credit-rationed farms, householdcharacteristics are proven to have a significant effect on output supply. Thisis evidence for a violation of separability between production and consumptiondecisions and thus lends empirical support to the existence of a marketimperfection. A counterfactual model estimated on all short-term credit recipientsdemonstrates that the willingness to pay is substantially higher forrationed than for non-rationed farm households.8. An ex-post evaluation of investment activities suggests that non-productiveinvestment rank high on the priority list of interviewed farmers. Residentialbuildings and automobile purchases are the two items with the largest shareof farm-individual investment expenses in the reporting period. The econometricinvestment analysis demonstrates that credit access is a significant factorof investment decisions of credit-rationed farmers. This supports the theoreticalprediction of a financial constraint model of investment behaviour andis consistent with the qualitative self-classification of respondents. Furthermore,the analysis substantiates the evidence that subsidised credit funds arepartly diverted to non-productive purposes. In various specifications of thecredit-investment relationship, the marginal effect of credit on productive investmentis clearly smaller than one. Based on a cubic Tobit estimate of theinvestment function, the mean of the farm-individual marginal effects is at.53 on average. Every second borrower invests less in productive assets thanhe borrows. Only 1.6 percent of the selected respondents with positive investmentdisplay farm-individual credit effects larger than one. Over the observedrange of credit volumes, the marginal effect increases with an increasing credit volume. However, the results do not support the view that investmentis positively related to farm size.In summary, the analysis provides evidence that credit rationing is a relevantphenomenon in rural Poland. A significant fraction of borrowers could substantiallyincrease their productivity if access to working capital were improved.However, the examination of long-term loans revealed that farmers often preferthe investment in non-productive assets to growth investment. Credit rationinghence is unlikely to be the ultimate constraint for modernisation and structuralchange in the Polish farm sector.Government intervention in its current form has clearly failed to eliminate creditrationing, and the targeting of state sponsored funds turns out to be rather dubious.An alternative government policy should aim to improve the general creditworthinessof prospective borrowers and address the causes of loan default inthe past that led to a poor reputation of certain borrowers. Policy action that improvesthe access to working capital should be given priority. One of the potentialside-effects of the introduction of direct payments under the CAP could beto relax exactly this constraint on working capital for currently credit-rationedfarm households.Beyond a further integration of theory and empirics in the area of New InstitutionalEconomics, an important focus of future research should be the developmentof analytical tools for practical policy advice on markets with pervasiveagency relations. With regard to Poland, political aspects of governmental creditmarket intervention as well as a more comprehensive analysis of the determinantsof structural change in the farming sector appear to be promising researchfields.