The financial crisis of 2008/09 and the national governments’ endeavours to stimulate the economy has rekindled interest in the Japanese experience with fiscal stimulus plans. Like in the ongoing economic crisis, many elements of the Japanese crisis and fiscal stimulus responses share similarities. During the 1990s, a period frequently referred to as the "lost decade", economic growth in Japan declined sharply to an average of 1.2 percent from an average growth rate of 3.9 percent in the decade earlier. The economic slowdown was precipitated by a bursting of the asset and credit bubble as the stock market declined by 41 percent and credit flow declined by 71 percent between 1989 and 1991. In response, the Japanese government introduced numerous stimulus packages which were continued over the course of the decade. Yet, despite large and repeated fiscal stimulus packages, the cost effectiveness of these packages has been questionable. Growth remained stagnant amidst deflationary pressures and public deficits and debts rose rapidly reflecting both tax revenue slowdown and the increase in government expenditures.
We contribute in this paper to the debate on whether government expenditures are effective in stimulating economic activity during times of crisis by exploiting a rich dataset of regional government expenditures in Japan during the 1990-2000 period to estimate from within-prefecture variation the multiplier effects that government investment and government consumption expenditures have on output. The use of regional data allows us to circumvent important identification issues that arise in country-level VAR analysis of the effects that fiscal policy has on output due to the non-passiveness of monetary policy as well as other "big shocks" that are often difficult to control for. We deal with these issues by using panel fixed effects regressions that account for both prefecture-specific unobservables as well as time-specific shocks that are common across prefectures in a given year.
Based on the within-prefecture variation of our regional data we find that the output multipliers of government expenditures are generally smaller than one. While government investment expenditures did have a significant positive effect on output, government consumption expenditures had either an insignificant ouput effect or, if government spending was on personnel, a significant negative effect. We also distinguish in our empirical analysis between government investment spending undertaken by the different levels of government administration. There we find that projects undertaken by the city government were about twice as productive as projects undertaken by the central government. While this points towards decentralized government investment being more effective in stimulating economic activity than centralized government investment, the multiplier on city level investment was still smaller than one.
To explain the small magnitude of the multipliers we examine the response of private consumption and private investment to government expenditure shocks. We find that both private consumption and private investment declined substantially as a consequence of increases in government consumption expenditures. Increases in government investment expenditures on the other hand led to substantial crowding out of private consumption but had a positive though insignificant effect on private investment. Hence, on the demand side we find that government expenditures were associated with the crowding out of private demand, both in terms of private investment as well as private consumption.
We also document using the regional data a strong decline in the government expenditure multiplier over time. While over the course of the 70s and 80s a 1 percent increase in government expenditures was associated with an increase in output by about 0.4 percent, an expenditure shock of similar magnitude increased output during the 90s by only about 0.1 percent. Hence, the size of the government expenditure multiplier during the 90s was only about one-fourth of the size of the multiplier during the 70s and 80s. This decline in the multiplier fits well with the decline in the marginal product of public capital that we document to explain why government expenditure increases had relatively small output effects in Japan during the 1990-2000 period from a supply side point of view. The decline in the marginal product of public capital resonates Hayashi and Prescott (2002) who argue that the "lost decade" in Japan was associated with a strong decline in total factor productivity.
There exists a fairly large empirical macro literature that has estimated the effects of government expenditures on output using VAR techniques. Our paper differs from these studies in mostly three aspects. First, we use prefecture data on total public investment as well as local government expenditures to calculate the multiplier. Local governments play an important role in Japan as they comprise 60 percent of spending and an even larger share in public investment. About 80 percent of public works projects were implemented by local government and about 65 percent were financed by the local government during the nineties. The availability of a rich dataset allows us to explore various dimensions of fiscal policy including the different types of spending and regional allocation during the nineties. Secondly, we use Generalized Method of Moments (GMM) to estimate the multiplier effects, which allows us to obtain consistent estimates even if government expenditures are affected on impact by changes in the regional economic environment. Finally, by extending the analysis of the marginal productivity of capital beyond the so-called lost decade", we complement the analysis of multiplier effects with the longer-term productivity of public investment spending.
The remainder of the paper is organized as follows. Section 2 provides a background on the fiscal stimulus packages and a review of the literature on the effectiveness of public spending in Japan. Section 3 describes the empirical strategy and the data. Section 4 presents the results on the estimation of multiplier effects of public spending. Section 5 assesses in further detail the possible reasons for the size of the multipliers with a focus on crowding out effects and the declining multiplier over time. Section 6 concludes.
