Ebook Adjustment to Target Capital, Finance, and Growth
In countries with well developed financial markets, capital is supposed to flow quickly to where it is expected to be most productive. Undeveloped financial markets, on the other hand, should slow down efficient capital reallocation and therefore reduce the speed of adjustment to supply and demand shifts and the rate of economic growth.
One way to verify this claim is by examining whether countries with better developed financial markets experience faster growth in industries expected to do well compared to industries anticipated to decline. Such an analysis requires a proxy for expected future growth opportunities at the industry level. One approach, following Rajan and Zingales (1998), is to use data from a country where one expects available industry data to closely reflect the relevant latent industry characteristics. For example, Rajan and Zingales use US investment and cash flow data to construct a proxy for industry characteristics that translate into greater demand for external finance independently of the country where the industry is located. The flexibility of US markets and the availability of reliable industry data make the US the natural benchmark. The high US levels of financial development also make it the place to start searching for measures of the expected future growth opportunities of industries.
Even the most developed financial markets can only reallocate capital in response to growth opportunities that could have been anticipated by businesses and financial markets. Tests of the role of finance for the speed of capital reallocation therefore require measures of expected growth opportunities. As investment decisions are forward looking, the growth of an industry’s capital stock should be an indicator of anticipated opportunities in economies with well-functioning financial markets. Industry capital growth in the US will therefore be one of our indicators of expected growth opportunities.
US data will at best be a noisy proxy for industry characteristics in other countries and such measurement error may lead to biased estimates of the effect of financial markets on capital reallocation and growth. While this was to be expected, it may be surprising that measurement error need not result in attenuation bias. Instead, the bias could lead to overstatements of the role of financial development for industry growth. This is because proxying country specific investment opportunities using US data leads to a measurement error that is country specific. Consider estimating the effect of financial markets on growth of industries with good investment opportunities country by country. Using US data to obtain a measure of opportunities in other countries will of course lead to attenuation bias. This bias will be less severe for countries experiencing similar demand and supply shifts as the US however. For example, there will be no bias at all for countries experiencing the exact same shifts as the US. For countries with idiosyncrasies that are unrelated to the US, however, attenuation bias could be severe. In this case, we will underestimate the effect of financial markets on growth in industries with good prospects less in economies that resemble the US. If economies resembling the US also have similar (high) levels of financial development, then a least squares approach will tend to exaggerate the role played by financial development for growth in industries with good investment opportunities.
Consistent estimation of the effect of cross-country differences in financial development on the growth of industries with good prospects therefore requires a measure of investment opportunities that does not reflect US idiosyncrasies, nor the idiosyncrasies of another country. To obtain such a measure we use industry value added growth in all countries with available data, except the US, to predict US capital growth. This additional (international) proxy of investment opportunities does not reflect the idiosyncrasies of any particular country by construction, but turns out to be closely related to US industry capital growth. We can therefore obtain consistent estimates of the effect of financial development on industries with good prospects by using .the international indicator of investment opportunities as an instrument for the proxy based on US data.
Our empirical results indicate a statistically and economically significant effect of financial market development on the growth of industries with good investment opportunities. This effect remains stable when we account for other determinants of industry growth across countries (e.g. Rajan and Zingales, 1998; Claessens and Laeven, 2003). Moreover, when we address measurement error using our indicator of international investment opportunities, we find larger and more significant effects of financial development on growth in industries with good prospects. These results prevail when we eliminate low-income countries from our sample or when we account for the possibility that the speed of capital reallocation maydepend on income levels. We also find evidence that growth in industries with good prospects is more closely related to financial development than to institutions beyond the organization of financial markets. Financial development does therefore not appear to stand-in for other aspects of country’s institutional quality. Moreover, we show that our findings are not driven by feedback from the pattern of industry growth to financial development.
Our paper is closely related to the general literature on finance on growth (e.g. King and Levine, 1993; Levine, Loyaza and Beck, 2000; Demirguk-Kunt and Maksimovic, 1998; see Levine, 2004, for a survey) and particularly Rajan and Zingales’ (1998) and Fisman and Love’s (2004a) work using cross-industry cross country data. Rajan and Zingales examine whether better developed financial markets translate into a comparative advantage in finance dependent industries. Our analysis differs in that we focus on the role of finance for the speed of capital reallocation towards industries with good prospects. Fisman and Love (2004a) examine the link between financial markets and interindustry resource reallocation using US sales growth as a proxy of growth opportunities. We measure growth opportunities anticipated by businesses and financial markets using investment data. The forward-looking nature of investment decisions should make them reflections of prospects. Sales (or value added) growth also depends on the effect of unexpected demand changes on prices. Sales growth should therefore be a noisier proxy of expected growth opportunities than measures based on investment data, a conjecture supported by our empirical evidence. Most importantly, we complement our US based measure of industry investment opportunities by a measure that is not affected by the idiosyncrasies of any particular country.
This allows us to estimate the effect of cross-country differences in financial development on the growth of industries with good prospects using a (consistent) instrumental variables approach. Fisman and Love (2004b) and Wurgler (2000) also use industry-country data to examine the role of financial development for investment and growth. Fisman and Love regress cross-industry value added growth correlations calculated for all possible country pairs on levels of income and financial development to show that correlations are greater the closer countries’ levels of financial development. Wurgler (2000) estimates the responsiveness of investment growth to value added growth and shows that this elasticity is greater in countries with better developed financial markets.
The remainder of the paper is structured as follows. Section 2 presents a theoretical framework linking financial development, capital reallocation, and industry value added growth. Investment decisions are driven by expected productivity growth, demand shifts, and global prices. Financial development (potentially) speeds up the adjustment of industry capital stocks towards their target values. We then use the theoretical framework to discuss estimation. This allows us to illustrate both the usefulness and the potential biases of using characteristics of US industries (or another country) to proxy for industry characteristics in other countries. We conclude with a description of our approach to dealing with country-specific idiosyncratic noise. Section 3 describes the sources and main features of our data. Section 4 presents our main empirical results. We first show that financial markets foster growth of industries with good prospects using our US measure of expected investment opportunities. We then present instrumental variable results that identify the effects of global expected growth opportunities. Section 5 contains additional evidence and further sensitivity checks. Section 7 summarizes and concludes.
Download
PDF Ebook Adjustment to Target Capital, Finance, and Growth
Posted in :
