Ebook Forecasting the Recent Behavior of U.S. Business Fixed Investment Spending: An Analysis of Competing Models
Empirical models of U.S. business fixed investment spending have a long tradition. This is not surprising, given the crucial role of investment spending in determining both long-term growth and fluctuations in aggregate activity at business-cycle horizons. The empirical literature has considered a number of different models of investment spending. Among the most popular models are the Accelerator (Clark, 1917; Chenery, 1952; Koyck, 1954), Neoclassical (Jorgenson, 1963; Hall and Jorgenson, 1967; Jorgenson, 1971), Tobin’s Q (Tobin, 1969), and Cash-Flow (Meyer and Kuh, 1957; Duesenberry, 1958; Grunfeld, 1960). A common approach in the empirical literature is to include one or more of these conventional models in a ‘horse race’ designed to identify the model (or models) that best explains U.S. business fixed investment spending over a particular period.
A partial list of empirical studies that run horse races includes Jorgenson and Seibert (1968), Jorgenson, Hunter, and Nadiri (1970a,b), Bischoff (1971), Clark (1979), Bernanke, Bohn, and Reiss (1988), Barro (1990), Blanchard, Rhee, and Summers (1993), Oliner, Rudebusch, and Sichel (1995), Kopcke and Brauman (2001), and Tevlin and Whelan (2003). Many of these papers feature horse races comparing the out-of-sample forecasting ability of competing models, owing to the widely held belief that tests of out-of-sample predictive power are the most stringent tests of a model’s reliability, especially insofar as out-of-sample tests guard against structural instabilities and data mining. Furthermore, given the interest of policymakers in forecasting U.S. business fixed investment spending, out-of-sample tests constitute a relevant test design for the examination of forecasting models (Poon and Granger, 2003).
Out-of-sample horse races have been run over different periods in the extant literature, ranging from the 1960s to the mid-1990s. In the present paper, we run out-of-sample horse races involving a number of forecasting models of U.S. business fixed investment spending over the more recent 1995:1-2004:2 out-of-sample period. This period witnessed an investment ‘boom’ that was at the center of the longest economic expansion in U.S. history, as well as an investment ‘bust’ that contributed significantly to the economic recession of 2001. We view this volatile period of investment spending as a useful laboratory for comparing different forecasting models of U.S. business fixed investment spending, so that the present paper represents an important update of the horse race literature on U.S. business fixed investment spending.
We consider six forecasting models of U.S. fixed private nonresidential investment spending growth at forecast horizons of 1-4, 6, and 8 quarters, with each model utilizing a different explanatory variable (or variables). Each of the first four forecasting models uses a variable suggested by one of the conventional models cited in the opening paragraph: real business output (Accelerator); real business output divided by the real user cost of capital (Neoclassical); the market value of capital relative to its replacement cost or average Q (Tobin’s Q); real profits (Cash-Flow). The fifth forecasting model follows Barro (1990) and uses real stock prices in place of average Q, as Barro (1990) argues that stock prices are a potentially better measure of the market’s assessment of the future profitability of investment projects than average Q.
The sixth forecasting model of U.S. fixed private nonresidential investment spending growth is motivated by the recent work of Lettau and Ludvigson (2002). Building on the modern Q theory of investment spending [see Abel (1982) and Abel and Blanchard (1986)], Lettau and Ludvigson (2002) argue that variables demonstrating predictive ability with respect to the equity risk premium (excess stock returns) should also have predictive ability with respect to business fixed investment spending. The sixth forecasting model includes three of the excess stock return predictors the term spread, default spread, and relative short-term interest rate that Lettau and Ludvigson (2002) find to have in-sample predictive ability with respect to U.S. business fixed investment spending growth.
Posted in :
