Microeconomic Risk Management and Macroeconomic Stability

In the traditional hedging literature, the two markets in which hedgers trade are spot and futures markets. The trader’s position in the spot market is generally considered as given. According to Johnson (1960), hedging can be meaningfully defined only if the spot market is regarded as the trader’s primary market. The futures market is used solely to counterbalance an existing position in the spot market. Speculators, in contrast, do not have a commitment in the spot market. They take on risk in futures markets in order to profit from expected price changes. The hedger synchronizes his trading activities in spot and futures markets in order to reduce spot risk. In the literature this approach to hedging is labeled risk reduction concept. Risk reduction will be achieved if spot and futures prices move more or less in parallel. If prices are perfectly correlated, risk is abolished, since losses in one market are perfectly offset by profits in the other market. However, as Hardy and Lyon (1923) point out, any divergence from perfect correlation results in an imperfect hedge. The less futures and spot prices move in parallel, the more imperfect the protection offered by hedging is. According to Kobold (1986), spot and futures prices generally do not move exactly in parallel. In fact, futures and spot markets are separate markets. Even speaking of a single spot market may be misleading, since, in general, most commodities are traded in many different places. The futures market, on the contrary, is generally highly centralized. Telser (1986) points out that each futures contract is a perfect substitute for another futures contract with the same maturity. If spot and futures prices do not move exactly in parallel, hedges end up with a profit or loss. Hence, if the motive for hedging is the elimination of spot risk, spot and futures prices, not moving in parallel, prevent complete risk reduction and are therefore unfavorable.

Most recent models on optimal hedging deal with exporting firms facing price or exchange rate risk. In order to hedge the spot commitment, firms go short in futures contracts.¹ This hedging literature, dealing with exporting firms hedging short, unequivocally suggests a negative relation between backwardation and the size of the optimal short hedging position.² In sum, the literature suggests that if the futures market is characterized by backwardation (contango), it is optimal for the short hedger to underhedge (overhedge), where underhedging (overhedging) means choosing a futures position smaller (larger) than the initial spot commitment. In the absence of backwardation or contango, the firm hedges fully, and therefore chooses the futures position to be the same size as the spot position.³ Hence, an increase in backwardation should, ceteris paribus, reduce the trading volume of hedgers in short futures contracts.

The hedging model introduced in this chapter is an extension of the expected utility approach in Chap. 2 The importing firm’s hedging problem here is almost identical to the one before. The only difference is that this model allows for basis risk. This is important with regard to the definition of backwardation applied. While in the previous chapter backwardation is defined as the difference between the expected spot price and the current futures price (i.e., \(\tilde{{e}}_{1} - {f}_{0}\)), here, backwardation is defined as the difference between the expected futures price and the current futures price (i.e., \(\tilde{{f}}_{1} - {f}_{0}\)). Note that these two definitions of backwardation are equal in the absence of basis risk (i.e., if \(\tilde{{e}}_{1} =\tilde{ {f}}_{1}\)). However, aside from basis risk, the model framework is quite different, since the analysis in this chapter is based on the mean-variance concept. Nevertheless, this approach can be regarded as an extension, since mean-variance models are generally not in conflict with expected utility models.¹ On the contrary, mean-variance models have several attractive properties that may add additional insights.

This chapter deals with the role of corporate risk management for macroeconomic stability. Firms’ balance sheets and the financing-investment relationship are at the center of this study. The interrelation of firms’ balance sheets and investment has been extensively investigated in connection with monetary policy transmission and, in particular, in connection with the balance sheet channel. Bernanke and Gertler (1990, 1995), Bernanke and Lown (1991), Calomiris and Hubbard (1990), Gertler and Gilchrist (1994), and Onliner and Rudebusch (1996) model investment as being sensitive to current cash flows and net worth.¹ A decrease in a firm’s cash flow and, hence, in a firm’s net worth will decrease its ability to borrow. This leads to investment contraction. An initial monetary shock, which worsens credit market conditions, can therefore result in large cycles as described by the financial accelerator.² The role of balance sheets in currency and financial crises are also well recognized.³ Mishkin (1998, p. 13) for example states that: “(...), there is another factor affecting balance sheets that can be extremely important in precipitating financial instability in emerging market countries that is not operational in most industrialized countries: unanticipated exchange rate depreciation or devaluation. Because of uncertainty about the future value of the domestic currency, many nonfinancial firms, banks and governments in emerging market countries find it much easier to issue debt if the debt is denominated in foreign currency. (...) With debt contracts denominated in foreign currency, when there is an unanticipated depreciation or devaluation of the domestic currency, the debt burden of domestic firms increases.”

This chapter studies nonlinearities and complexity in currency futures markets. First, the impact of price changes on trading volume is empirically investigated using linear vector autoregression analysis and nonlinear logistic smooth transition regression analysis. Second, the empirical findings regarding nonlinearities in traders’ behavior, together with economic theory concerning arbitrage pressure and noise trading, are modelled in a cusp catastrophe model. There is a large body of literature dealing with nonlinearities in financial markets.¹ These studies generally analyze nonlinearities in prices due to inefficient arbitrage and the existence of noise traders.² The empirical study in this chapter differs considerably from the one chosen in the studies mentioned above, since it focusses on nonlinearities in the responses of the quantity of trading volume to price changes. The empirical investigation follows Röthig and Chiarella (2007), and applies the logistic smooth transition regression (LSTR) model to investigate the impact of changes of currency futures settlement prices on the trading positions of futures traders. Smooth transition regression models have been widely used in a range of different fields of research, including stock market returns, exchange rates and interest rates³, monetary economics⁴, GDP growth⁵, business cycles⁶, and for modelling phenomena like El Niño.⁷

A hedger is a trader who simultaneously holds positions in spot and futures markets in order to reduce spot exposure. However, he does not necessarily minimize the initial spot risk. The minimization of risk is just one single possible outcome from a wide range of potential hedging strategies. Nevertheless, hedging less than the initial spot commitment does not mean that the hedger turns into a speculator. This is because he still holds positions in both markets, with the result that his overall exposure to risk is smaller than if he would only trade in the spot market. Risk is reduced, but not eliminated. The microeconomic part of this thesis focuses on the determinants of firms’ optimal hedging strategies. The impact of price expectations, risk aversion, and hedging costs are particularly important. If hedgers expect spot prices to move in their favor they will be less willing to hedge, since potential returns in the spot market are offset by losses in the futures position. In the presence of hedging costs, the overall profit of the hedged position would be negative if earnings and losses in spot and futures markets were perfectly balanced. Hence, under risk neutrality, companies will not hedge unless there are other incentives to hedge that outweigh the costs. These incentives include taxes, bankruptcy costs and underinvestment problems among others. The models presented in the microeconomic part of this thesis assume a priori that the hedging firm is risk averse.