Issues Facing the Japanese Economy in 2013 (January 2013)

Economic Forecasts and Uncertainty from an Economic Growth Point of View

MORIKAWA Masayuki
Vice Chairman & Vice President, RIETI

The real growth rate of the Japanese economy for the July-September quarter of 2012 was -0.9% (an annualized rate of -3.5%). Economic growth decelerated from 1.4% for the January-March quarter to 0.0% for the April-June quarter, and experienced significant negative growth in the July-September quarter. Many believe that the Japanese economy has already entered a recession. In this environment, the government was overturned at the end of 2012 for the first time in three years, and many changes in economic policy are expected.

The consumption tax and the economic outlook for the next fiscal year

With the enactment of an act to revise the Consumption Tax Act in August 2012, it was decided that the consumption tax rate would be increased to 8% in April 2014. The so-called consumption tax hike law, however, states that the consumption tax rate will be raised only if the economy experiences an upturn. More specifically, supplementary provisions of the law stipulate that appropriate measures, including the suspension of the planned increase in the consumption tax rate, shall be taken if necessary after comprehensively reviewing the economic situation by checking economic indicators such as nominal and real economic growth rates and price trends. Therefore, economic trends in fiscal 2013 (April 2013 - March 2014) could have a significant impact on the country's medium- to long-term fiscal management.

Private think tanks have already announced their economic forecasts for the forthcoming fiscal year, and it appears many of them expect Japan's economy to improve gradually along with a moderate recovery of the global economy, although the numbers vary depending on the institution. They think that, domestically, continuing demand from post-quake projects and last-minute demand before the hike in the consumption tax rate will have a positive effect on the economy. As risk factors, they cite the "fiscal cliff" in the United States, the escalation of the European crisis, anti-Japanese movements in China, and the appreciation of the yen, etc (note 1). Meanwhile, although the government economic outlook is usually submitted to and approved by the Cabinet in late December, it was decided that this would be postponed until after the new year, as is the case with the national budget, because of the inauguration of a new government upon the general election in December. Expected changes in the basic policy for managing the economy resulting from the change of government, coupled with the aforementioned circumstances surrounding the planned tax hike, conceivably make the preparation of the economic outlook extremely difficult.

Accuracy of the economic outlook

The government's economic outlook has an important role, serving as the basis for the overall economic management including tax revenue estimation and expenditure plans. However, as this economic outlook is prepared based on the assumption that various factors—global economic trends, crude oil prices, exchange rates—will remain steady throughout the subject fiscal year, the actual economy can deviate significantly from the outlook, depending on movements in these given conditions. Particularly in recent years, we have witnessed the occurrence of a series of large-scale shocks that were unpredictable at the time of the forecast, such as the global financial crisis and the Great East Japan Earthquake.

Comparing the government's forecasts to the corresponding actual figures for past years, we can see that actual economic performance tended to show a significant deviation from the forecast when there was a negative exogenous shock, as has been the case with the economic crisis in Asia, the bursting of the information technology (IT) bubble, the collapse of Lehman Brothers, and the Great East Japan Earthquake. To begin with, the government's forecasts have an average error of ±2 percentage points or greater for both real and nominal growth. If we take the actual growth rate in the preceding fiscal year (or the average growth rate for the preceding three years) as a forecast value for the current fiscal year, forecast errors for real gross domestic product (GDP) growth are slightly smaller than those of the government's forecasts. On the other hand, in the case of nominal GDP growth, growth forecasts based on actual growth in the past are more accurate than those by the government (see Figure) (note 2). Relatively large errors observed in the government's nominal GDP growth forecasts are attributable to its consistent tendency to overestimate the inflation rate. This suggests the possibility that the government has had an optimistic bias, faced with the compelling need to bail the nation out of deflation, a long-standing policy challenge.

Figure: Errors in GDP growth rate forecasts for FY1990 - FY2011 (RMSE)Figure: Errors in GDP growth rate forecasts for FY1990 - FY2011 (RMSE)

Many studies formally examined the ex post facto accuracy of economic forecasts. As an example of analyzing the Japanese government's economic outlook, I would like to cite a study by Ashiya (2007). Examining the records of forecast and actual figures over a period of 22 years, the study found that the government's year-ahead forecasts for real GDP growth had a 0.7 percentage point upward bias on average. Meanwhile, Frankel (2011) conducted an empirical analysis of biases in official predicted values for real economic growth rates and the budget balance of the governments of 33 countries. The results showed that government economic forecasts have an upward bias, and that the longer the prediction period, the greater the bias (note 3). On that basis, Frankel argues that the excessively optimistic government economic forecast is the reason for excess fiscal deficits and, particularly, the failure to create a budget surplus when the economy is strong.

Although the need for highly accurate economic forecasts is great amidst the increasingly uncertain outlook for the economy, there is significant uncertainty in the government economic outlook as well.

Uncertainty and economic activities

Uncertainty over the future of the economy has been heightened since the global financial crisis. It is noted that this economic uncertainty has a large impact on the behavior of companies and households. Bloom et al. (2012) have analyzed the impact of uncertainty on the economy using the dynamic general equilibrium model, and showed that an increase in uncertainty makes it best for companies to wait until prospects become clearer, and drives down employment, investment, and production significantly. They also show that uncertainty reduces the redistribution of resources in the economy as a whole and lowers the rate of productivity growth. In their analysis of uncertainty in the United States and the United Kingdom, Leduc and Sill (2012) argue that uncertainty over the future of the economy has a negative impact on the macro economy similar to that of a decline in aggregate demand, and results in a higher unemployment rate and a deflationary trend. These studies indicate the importance of reducing uncertainty and the government not creating additional uncertainties.

Japan saw a change of government for the first time in three years as a result of the general election at the end of 2012. Many policies changed when the Democratic Party of Japan took over the reins of the government from the Liberal Democratic Party, and the same is expected with the government changeover this time. Discussing the right or wrong of individual policies is not the purpose of this column. However, I would like to point out that not only the content of the policies but also their instability will have an impact on economic activities, as companies and households formulate their plans for business management, consumption, and asset management on the assumption that the existing institutions and policies will continue. For example, Fatas and Mihov (forthcoming) show in their empirical analysis using panel data for 93 countries between 1960 and 2007 that frequent changes in fiscal policy have a significant negative impact on economic growth. As an indicator of policy volatility, they use changes in fiscal spending attributable to political factors, or, more specifically, the degree of change in government consumption that cannot be explained by cyclical effects. Then, after taking into account various factors that could have an impact on economic growth, they estimate that for a one-standard deviation increase in policy volatility, the long-term economic growth rate will decline by 0.7 percentage point or more on an annual basis. Based on this result, they argue that the government should be careful when changing policies, taking into account the long-term impact of policy volatility (note 4). In a mature economy like Japan, an economic growth rate of 0.7% is quite significant, and there is no individual growth policy with such a large quantitative effect. Although drastic changes in policies have strong appeal politically, we must pay attention to the risk involved as they could have a negative impact on economic growth.

These empirical studies in Japan and overseas suggest the importance of building a stable and sustainable policy framework for the national economy based on cooperation between the ruling and opposition parties, particularly at a time when fiscal sustainability is in danger.

December 28, 2012
Footnote(s)
  1. ^ The OECD Economic Outlook has a relatively cautious view on Japan's economic growth rate, forecasting 0.7% for 2013 and 0.8% for 2014.
  2. ^ Forecast errors discussed here are root mean square errors (RMSE). For the government economic outlook, gross national product (GNP), not GDP, is used until fiscal 1993, and chain-linked real GDP is used after fiscal 2005. For actual values, corresponding series are used. However, the calculation is made here by using the retroactive series of the Annual Reports on National Accounts instead of real-time data at the time of forecast.
  3. ^ The upward bias in GDP growth forecasts is estimated to be 0.4 percentage points for one year, 1.1 percentage points for two years, and 1.8 percentage points for three years.
  4. ^ In its latest World Economic Outlook, the International Monetary Fund (IMF) also points out the negative impact of uncertainty derived from policies on economic growth.
Reference(s)
  • Ashiya, Masahiro (2007). "Forecast Accuracy of the Japanese Government: Its Year-Ahead GDP Forecast Is Too Optimistic." Japan and the World Economy, Vol. 19, No. 1, pp. 68-85.
  • Bloom, Nicholas, Max Floetotto, Nir Jaimovich, Itay Saporta-Eksten, and Stephen J. Terry (2012). "Really Uncertain Business Cycles." NBER Working Paper, No. 18245.
  • Fatas, Antonio and Ilian Mihov (forthcoming). "Policy Volatility, Institutions and Economic Growth." Review of Economics and Statistics.
  • Frankel, Jeffrey A. (2011). "Over-optimism in Forecasts by Official Budget Agencies and Its Implications." NBER Working Paper, No. 17239.
  • Leduc, Sylvain and Keith Sill (2012). "Uncertainty Shocks are Aggregate Demand Shocks." FRB San Francisco Working Paper, No. 2012-10.

January 21, 2013