chapter  10
19 Pages

Business investment, growth and crisis

ByANA - ROSA GONZÁLEZ , PHILIP ARESTIS AND ÓSCAR DEJUÁN

The current economic crisis was accompanied by a collapse in business investment.1 The recovery won’t happen until a massive wave of investment occurs. There is nothing new in these assertions. Economic downturns and upturns coincide with the deceleration and acceleration of business investment. But, is investment the cause or the consequence of the cycle and the crisis? This is the first question addressed in this chapter. The second consists in identifying the whole set of determinants of investment. This knowledge will help us to find the best economic policies to overcome the first great recession of the twenty-first century. Our chapter is an empirical study based on post-Keynesian theories that develops the seminal ideas sowed by Kalecki and Keynes after the Great Depression of 1929 (Kalecki, 1971; Keynes, 1936). From this theoretical standpoint the answer to the first question is “both”. On the one hand, business investment depends on the expected growth of demand that may be influenced (perhaps unduly) by the current growth of output. On the other hand, since it is an important part of aggregate demand, a fall in business investment is bound to depress the current level of output and its rate of growth. Keynes, in chapter 12 of the General Theory, makes clear that the main determinant of investment is the state of long-term expectations encapsulated in the animal spirits of the entrepreneurs. Expectations are the true independent variable of the Keynesian system (Eatwell, 1983). They refer, mainly, to the future demand of goods and services at normal prices (the difference between current and normal prices accounts for extra profits). Prediction errors result in overusing or under-using existing capacity which will speed up or slow down the investment determined by the pure accelerator mechanism that depends on the expected rate of growth and the optimal capital/output ratio. Harrod (1939) introduced the old principle of acceleration in a Keynesian growth model. Later on, it was made flexible enough to account for capacity utilization, financial constraints and so on (Chenery, 1952). The current degree of capacity utilization plays a prominent role in Kaleckian accumulation models (Lavoie et al., 2004). In our model what matters are the deviations of the current

degree from the “normal” level that entrepreneurs have in mind in their desire to maximize profits in the long run (Kurz, 1986). We shall identify normal capacity with the average of boom periods that is around 0.8. This means that entrepreneurs allow for 20 per cent of spare capacity to match the peaks in demand. Distributive variables and financial conditions are supposed to influence the level of investment and, most of all, the temporal implementation of the decisions based on the pure accelerator mechanism. Kaleckian and Marxian economists emphasize the importance of the rate of profits that can be proxied by the profit share (profits/income). A higher profit share, apart from spurring on investors, provides them with the internal funds necessary to purchase capital goods (Kalecki, 1971, 1968;2 Bhaduri and Marglin, 1990; Alexiou, 2010). The rate of interest is another distributive variable that indicates the cost of raising external funds. Other things equal, a fall in the real interest rate is supposed to speed up investment decisions. This justifies Keynes’s suggestion, in the last chapter of the General Theory, to keep the interest rate as low and stable as possible. The new macroeconomic consensus distrusts such a policy because of the risk of inflation. Arestis and Sawyer (2004) show the weakness of the theory behind this conclusion and prove that in a globalized economy inflationary pressures are checked by international competition. The prices of assets (not considered in the CPI) are the only exception. Cheap credit feeds speculative movements in the stock exchange and in the real estate market that, eventually, will damage business investment. Notice that this runs against Tobin’s q theory of investment (1969). In our opinion a rise in the price of shares makes external finance cheaper, but does not justify more investment. On the contrary, if people expect that the upward rally in the stock market will continue, they will sacrifice productive investment in favour of speculative purchases of assets. Credit is part of the financial conditions. As a matter of fact, when credit is introduced in an econometric model it becomes the only explicative variable of investment. Arguably, this confirms the post-Keynesian hypothesis of the endogenity of credit-money according to which credit is more the consequence than the cause of investment booms. Remember, however, that credit implies debt and an excess of bad quality debt may cause a downturn in investment and output. Kalecki (1937) introduced the principle of increasing risk to explain the upwards movement in interest rates due to excessive debt and leverage ratios. Minsky (1975, 1977) introduced the financial fragility hypothesis that explains endogenously the deterioration of debt: booms sow the seeds of depressions because they fill the market with bad quality debt. Entrepreneurs form their expectations of demand, prices, profits in an atmosphere of uncertainty (Demir, 2009). Post-Keynesian economists insist that we are dealing with radical uncertainty which rules out the probabilistic methods to deal with risk (Davidson, 1991). When uncertainty exacerbates, financial agents search for safe harbours for the savings they manage. They will invest them in real assets (like gold) and financial assets (like shares and bonds nominated in a currency free from depreciation). These harbours compete with productive investment that cannot emulate the yields of bullish asset markets (Stockhammer

and Grafl, 2010). A massive deviation of savings towards them will shrink productive investment and cause a problem of effective demand that may trigger a crisis. In our chapter strong increases in the price of gold, stock exchange indexes and the exchange rate will be taken as symptoms of uncertainty that discourage business investment. To be as rational as possible in a world of radical uncertainty, the society organizes proper “institutions” and economic agents try to rely on conventions (Littleboy, 1990; Runde, 1990; Crotty, 1992; Baddeley, 1999). An independent central bank, for instance, has been created to fix the inflation target and the official interest rate which is supposed to shape the whole structure of market interest rates. What matters for investment decisions is not the precise level of the interest rate, profit rate and so on, but the deviations from their conventional levels. The conventional level of these variables usually lacks a natural basis, given by technology, preferences and other factors of slow (and usually unidirectional) evolution (Dejuán, 2007). The conventional interest rate, to continue with the same example, is the one that has ruled in the recent past and people expect for the foreseeable future. A change in the official rate set by monetary authorities, if it lasts long enough, is bound to change the conventional value of the interest rate. This result demonstrates the importance of hysteresis in the economy. Most of the explanatory variables of investment will be treated in the same mood, i.e. deviations from the conventional level which may change from time to time. In this chapter we try to find out the determinants of private accumulation in 12 advanced economies from 1970 to 2008. (The availability of OECD data on fixed capital has determined the territorial and temporal size of the sample.) The second section focuses on a single economy (the USA) and a shorter period of time (the years prior to the crisis, starting in 1991). It’s a strategy to visualize the main relationships behind the econometric panel of data and connect them with the current crisis. Next we explain the econometric model for the whole series of data and evaluate the results. After trying several methods we have opted for the PCSE technique (Panel Corrected Standard Errors) (Beck and Katz, 1995). The use of panel data reflects the assumption that, in a globalized economy, countries share a common structure (Baltagi, 1995). Finally, we summarize the conclusions of the econometric study on accumulation and drag the policy strategies that might help to pave the way for the recovery of business investment.