Media Briefings

THE ECONOMIC IMPACT OF PUBLIC SPENDING IN A FINANCIAL CRISIS: Evidence from 1990s Japan on the key role of firms’ financial distress

  • Published Date: December 2014

The impact of government spending on output is smaller when firms are experiencing financial distress. That is the central finding of a study by Markus Brueckner and Anita Tuladhar, which is published in the December 2014 issue of the Economic Journal.

Analysing data on 47 Japanese prefectures during the 1990s, they find that firms’ financial distress significantly lowers the government spending multiplier: a 20% decrease in commercial land prices is associated with a 0.3 lower government spending multiplier (which would mean, for example, that instead of a 1 yen increase in local government spending leading to a 1 yen increase in output, output would only increase by 0.7 yen.)

The researchers conclude that the role of firms’ financing constraints should be taken into consideration when thinking about the effectiveness of government spending during a financial crisis.

The effectiveness of government spending in stimulating an economy during a financial crisis has been widely debated in recent years. Recent empirical research has indicated that economic slack (as measured by low economic growth or high unemployment) and the presence of the zero lower bound significantly increase the government spending multiplier while high government debt significantly reduces it.

The new study suggests that when thinking about how government spending multipliers are affected by a financial crisis there is another important aspect that needs to be taken into account: firms’ financial distress.

Research on incomplete financial markets (including a study by former Fed chairman Ben Bernanke) shows that reductions in firms’ net worth negatively affect firms’ borrowing capacity. These researchers therefore argue that decreases in commercial land prices are a plausible measure for firms’ financial distress.

Indeed, according to this measure, Japanese firms experienced significant financial distress during the 1990s. The price of commercial land declined in Japan by over 50% during the 1990s with a sample standard deviation across prefectures of around 20%. In the six largest city areas (Tokyo, Yokohama, Nagoya, Kyoto, Osaka and Kobe), the average decline was as large as over 80%.

The study uses this significant variation in the decline of commercial land price across Japanese prefectures to estimate state-dependence of the government spending multiplier on firms’ financial distress. (It also finds that firms’ financial distress lowers the government spending multiplier when using other measures of firms’ financial distress, such as reductions in bank lending and increases in non-performing loans.)

The finding that firms’ financial distress lowers the government spending multiplier is consistent with recent supply-side models of fiscal policy. In these models, government spending exacerbates firms’ financing problems due to a negative wealth effect.

This negative wealth effect arises because government spending has to be financed, either by higher taxes today or by higher taxes in the future. As firms find it more difficult to obtain external finance for starting new projects, they reduce demand for labour and capital. Consequently, the impact of government spending on output is smaller when firms experience financial distress.

Investigating the size of the government spending multiplier during a financial crisis, the researchers find that on average local government expenditures had a significant positive effect on output. The estimated local government spending multiplier is around 0.8 – that is, for each 1 yen increase in local government spending, output increased by around 0.8 yen.

Given the fact that during the 1990s, Japan experienced low economic growth (average GDP per capita growth was less than 1%) and the interest rate hit the zero lower bound towards the end of that period, the finding of a relatively modest effect of government spending on output may be surprising.

But as the study shows, the presence of economic slack and the zero lower bound (which increased the government spending multiplier) was counteracted by the presence of significant financial distress (which decreased the government spending multiplier).


Notes for editors: ‘Local Government Spending Multipliers and Financial Distress: Evidence from Japanese Prefectures’ by Markus Brueckner and Anita Tuladhar is published in the December 2014 issue of the Economic Journal.

Markus Brueckner is at the National University of Singapore. Anita Tuladhar is at the International Monetary Fund (IMF).

Disclaimer: The views in this paper are those of the authors alone and do not necessarily represent those of the IMF or IMF policy.

For further information: contact Markus Brueckner via email:; or Romesh Vaitilingam on +44-7768-661095 (email:; Twitter: @econromesh).