In this article we explain the fundamental factors determining investment decisions of firms, which comprise the investment function. By determining the structure of the investment function we can hypothetically estimate this and thus predict how much a given firm ought to be investing, given economic fundamentals. This is interesting because we could then aggregate such functions – i.e. add up each firm-specific investment equation – to get a measure of what total investment by private firms in the economy ought to be investing. By comparing this amount with actual levels of investment we can derive a measure of the investment gap.

Before we begin we define a few terms: investment is the building up of capital stocks, where a capital good is a good which is not used for consumption but makes future consumption possible through building new goods for the future. We can think of capital goods as being factories, tools, machines and computers which are used to build consumption goods. Depreciation is the wear and tear of capital goods which causes their value to fall – the more a saw is used, the blunter it becomes, and so it will need to be maintained or replace; this is depreciation.

A simple firm-specific investment function would have investment depending on expected future demand, the cost of capital and current/past levels of investment. The latter is the simplest variable to explain: if investment was low in the past then greater investment might be needed now to update the capital stock and meet current demand; higher past investment could mean that the capital stock is sufficiently built up, which would require less current investment, although it will also entail high levels of investment to prevent depreciation.

Fundamentally a firm will only invest in a project (e.g. a new factory) if it believes that the benefit of such a project (its future income stream) exceeds the cost. Hence investment depends on expected future demand with high demand meaning greater investment today to meet such demand, as well as the cost of capital.

The neoclassical model specifies that there should be no difference in the cost of capital between internal and external funds and so the level of internal funds a firm has should not matter for its investment decisions. Accordingly, if a firm has a profitable investment project then it would be able to obtain finance externally (e.g. from a bank) to fund such a project, and so wouldn’t need to have retained earnings[1] or other internal funds. However, asymmetric information, whereby borrowers have different information to lenders, may mean that there is. As a result if a firm has low internal funds (proxied by cash flow) then this could inhibit investment if external funds are more expensive or less accessible. This leads to pecking order theory – in contrast to the neoclassical theory – which tells us that internal funds are cheaper than external funds in financing firm investment.

This has important implications as profits (internal funds) are highly cyclical, and so if investment depends directly on the availability of profits then investment spending will be highly sensitive to fluctuations in economic activity. Furthermore, this could have implications for take-over activity, with firms with large internal funds taking over firms whose investment spending is constrained, and this being efficient; contrary to neoclassical predictions.

In our examination, this finding means that internal funds are an important component of firm-specific investment functions and so should be included in our regression. We can proxy this variable with either a profit variable or a cash-flow variable, however this leads to complications. Most micro studies confirm that financial variables such as cash flow help explain investment spending, with many authors attributing this to financial constraints. However the cash flow variable may be acting as a proxy variable for omitted expected future profitability rather than signalling fluctuations in internal finance availability and thus financial constraint. The term financial constraint means that a windfall increase in cash-flow, which conveys no new information about the firm’s investment opportunities, would produce an increase in investment spending.

Principally, a profit/cash-flow variable could be telling us information on the cost of capital, but it will also be telling us about information on expected future profitability. Therefore, we would be unable to distinguish – through econometric techniques – whether low firm specific investment was being caused by low expected future profitability, or a high cost of capital. In other words if we were to estimate a regression equation where investment depended on cash-flow and we found a significant negative result, then does this tell us that the cost of capital is high, or is it merely a signal telling firms that future demand will be low; both factors would inhibit investment.

We have examined the difficulties associated with including a cost of capital measure in our firm-specific investment function. Next we turn to issues associated with estimating future demand/profitability.

Tobin’s Q is used as this tells us the average profitability of investment (market value of firm capital/replacement cost of capital) and this equals the marginal decision (which can be calculated through optimisation and finding the shadow price, but which is not directly observable) under certain conditions according to Hayashi’s theorem. These conditions are that the production function has to be homogenous of degree 1 and the total adjustment cost function is homogenous of degree 1. The theory then tells us that average and marginal are the same and so Tobin’s Q tells us the marginal profitability of investment.

Under perfect assumptions the stock market valuation of a company captures all relevant information about expected future profitability and any significant coefficient on cash-flows after controlling for Tobin’s Q could be attributed to financial constraints.

However, Bond and Cummins (2001) show that due to asset market bubbles then stock market valuations may not reflect fundamentals and hence Tobin’s Q will not tell us exact investment profitability conditions. Hence cash flow could then provide additional information about expected profitability.

Bond, Klemm, Newton-Smith, Syed and Vlieghe find that cash flow becomes insignificant when controlling for expected profitability using analysts’ earnings forecasts, which shows that cash flow was proxying for expected profitability and so financial constraints may not be a significant impediment on firm investment. “Our principal conclusion is that, in line with standard economic theory, direct measures of expected future profits are very informative explanatory variables for the behaviour of company investment. In contrast, Tobin’s Q measures based on stock market valuations are much less informative, providing only marginally significant additional information after controlling for short term earnings forecasts. Moreover cash flow variables, which appear to be highly significant in reduced form models or in models which control for Tobin’s Q, become insignificant once we control for expected future profitability using analysts’ earnings forecasts…The stock market valuations contained in Tobin’s Q become completely uninformative in our empirical investment equations when additional variables like sales growth or cash stock are included together with expectations of short term profitability….The limited information in this measure of the average Q ratio is consistent with the presence of pervasive and persistent ‘bubbles’, or deviations between stock market values and the present discounted value of expected future profits. Alternatively this could indicate a failure of the Hayashi conditions – perfect competition, constant returns to scale and strictly convex adjustment costs – under which average Q is a sufficient statistic for investment rates.”

[1] Around two-thirds of investment is financed from retained earnings (Mayer, 1988).