Does governance create political risk? - The case of Investment decisions in agriculture.

Abstract

This thesis seeks to address how political risk influences investment in the agricultural sector in Norway. The agricultural sector is a particularly interesting case as it is highly affected by governmental intervention. Governance is aimed at establishing predictable and stable production terms, and an important question in this context is whether governance is able to reduce the overall risk faced by farmers. For example, the politically determined target prices for milk, aims at reducing the overall risk of dairy farmers, but they could also provide a false sense of security. A farmer in a free market will be exposed to price risk. A farmer in Norway will not be exposed to price risk in the same way because prices are more or less determined once a year in the agricultural negotiations. However, due to governmental intervention in Norway, the farmer will be more vulnerable to political risk than the farmer in the free market. One could therefore potentially argue that the overall level of risk is not reduced, but that the source of risk is different. In order to analyse the effect of political risk, I will look at aggregate investment in the agricultural sector and use two proxy variables for political risk. These proxies are the time-period of a parliamentary election and the time-period where a new government is elected. The theoretical foundation of the thesis will make use of the real option approach in analysing the effect of political risk. Here the value of the real option is linked to uncertainty about the future political framework and the ability to delay an investment decision until this uncertainty is resolved. If political risk is an important risk factor, the farmer may use the option to wait until the outcome of the election is known. If so, investment should be lower in the time before an election or in the time before an expected change in government compared to the time afterwards. The empirical part of this thesis is based on data from the Norwegian quarterly National Accounts (Nasjonalregnskapet) published by SSB, Farm Account data (Driftsgranskningstall) and the Aggregated Accounts for Agriculture (Totalkalkylen for Jordbruk) both published by NILF. These data sources will be used in order to analyse if variations in risk impact investment and the datasets contribute to the analysis in different ways. The Norwegian National accounts give quarterly data on investment that makes it possible to analyse if there is a drop in investment in the quarters before an election The Aggregated Accounts allow us to analyse whether the type of investment and the depreciation rate influence the option value. Farm Account data on the other hand; enable us to look at the effect of increased risk in one specific sector. The hypothesis that political risk influences investment has been tested with two different econometric specifications, given by the autoregressive conditional heteroscedastic (ARCH(1)) model and ordinary least square regression. The ARCH model specification measures if any of the dummy variables for risk influences the volatility of investment. The results indicate no conclusive evidence to support the hypothesis. The least square regression specification analyses whether political risk influences the investment level. While I do find support for the dummy variable on election of new governments on a 10 % significance level in the Farm Account dataset - the significance is weak and the results are not confirmed in other specifications. I find no conclusive evidence for political risk influencing neither the volatility of investment nor the investment level. This indicates that, given the data and the methods used in this study, political risk does not significantly influence producer behaviour in the agricultural sector. This could be interpreted as the agricultural sector being well regulated and that political conditions are relatively stable across different and changing governments, as well as over different party constellations in Parliament.