RDP 2005-12: Financial Constraints, the User Cost of Capital and Corporate Investment in Australia 1. Introduction

The pace and pattern of business investment in fixed capital are central to our understanding of economic activity. Business investment is one of the key determinants of an economy's long-term growth rate and also plays a pivotal role in explaining business cycle fluctuations. And yet, despite extensive empirical research, it has been hard to clearly identify a role for some of the factors that are believed to drive capital spending.

This at least partly reflects an ‘aggregation problem’ – the drivers of capital spending can vary a lot in terms of both their magnitude and timing across different industries and markets. To circumvent this problem, this paper uses panel data on individual firms to model investment spending in Australia. The use of panel data has several advantages over aggregate data. First, by introducing cross-sectional variation, panel data can help to minimise the simultaneity problem that often occurs in macro studies of investment. For example, policy-makers, in anticipation of future excess demand, might raise interest rates at the same time that investment demand increases. All else equal, this would result in a positive relationship between interest rates and investment, in contrast to the predictions of neoclassical theory. Second, cross-sectional variation allows the relationships between investment and its determinants to be measured using fairly short time series. Finally, panel-data analysis allows us to determine whether changes in financing costs are more important for certain types of firms or industries. By recognising firm and industry heterogeneity we can better understand the monetary policy transmission mechanism.

This paper estimates an error correction model (ECM) of corporate investment. The ECM framework has the relative advantage of modelling both the short-run dynamics and long-run determinants of investment, such as real sales growth and the user cost of capital. To the best of my knowledge, this is the first time this has been done using Australian panel data. As well as long-run determinants such as real sales growth and the user cost of capital, a vast literature suggests that financial factors are also important in explaining short-run fluctuations in investment.^[1] Transaction costs and/or information asymmetries induce a cost premium that makes external finance an imperfect substitute for internal finance (Jensen and Meckling 1976; Stiglitz and Weiss 1981; Myers and Majluf 1984). When capital markets are imperfect, some firms may be ‘financially constrained’ in the sense that they are unable to raise enough external funds to meet their desired level of investment spending. If this is the case, in the short run, the level of internal funding could significantly affect investment.

Including financial factors as short-run determinants is also important for understanding the monetary policy transmission process: if capital markets are imperfect and firms face additional costs of raising external finance, then interest rate changes may affect capital spending by influencing the amount of funds available. This effect can occur over and above the direct effect of interest rate changes on the relative price of capital goods and hence, business investment. It has become common practice to refer to these two channels of monetary policy as the ‘credit’ and ‘interest rate’ channels.

A finding that internal funding has a significant effect on firm-level investment can be taken as evidence of financial constraints. In principle, financially constrained firms should display greater sensitivity of investment to cash flow than unconstrained firms. However, internal funding could be relevant for investment because it also provides information about future investment opportunities (Bond et al 2004). To control for this, I also estimate a Q-type model in which the ratio of the market to book value of a firm's equity (Q) plays a key role in explaining its investment. This is the approach used in earlier Australian studies (Shuetrim, Lowe and Morling 1993; Mills, Morling and Tease 1994; Chapman, Junor and Stegman 1996).

The paper is organised as follows. Section 2 begins with a discussion of the role of financial constraints in influencing investment. Section 3 discusses the panel data used and compares measures of investment, sales, the user cost of capital and cash flow at both the micro and macro levels. A model is then estimated in Section 4 to examine the determinants of corporate investment in Australia. This section also outlines the key results and makes some comparisons with the results of other international studies. Section 5 concludes.

Footnote

For surveys of the literature, see Blundell, Bond and Meghir (1992), Chirinko (1993), Schiantarelli (1996), Hubbard (1998) and Bond and Van Reenen (2003). [1]