Raising Growth in Japan: What role can private consumption and investment play?

*As per the author's request, this transcript is not for quotation.

Kiichi TOKUOKA
The Japanese economy was hit extremely hard by the recent economic recession even though the Japanese banking sector was not directly affected by the housing problem. It is obvious that Japan needs to raise medium-term growth going forward. In terms of domestic demand, there are a couple of things that Japan can do. Japan can boost private consumption, capitalize on growth in emerging Asian economies in the form of exports, and invest in the service sector as a buffer against external shocks.

For private consumption, I will outline why private consumption has stagnated and what the possible policy options are to boost it. For investment, we will explore what has been driving investment in recent years and what policy options can be implemented to facilitate post-crisis adjustment.

Japan's private consumption share as a percentage of GDP has been lower than most G7 economies, though it is not too low by OECD standards. Consumption growth has been lagging overall economic growth since 2000. Looking at consumption and disposable income, there is some room for expanding private consumption as well as boosting consumption growth. There is a high relation between increases in consumption and disposable income. Economic regressions suggest that the ratio of elderly people in Japan also contributes to moves in private consumption.

Two major components of disposable income are wages and property income. However, both wages and property income have been very weak in Japan. My colleague Martin Sommer has done work on this, and his research suggests that an increasing share of non-regular workers, lower labor productivity and low globalization are contributing to the declining labor share in Japan. According to the OECD, protection for permanent workers is not particularly strong in Japan. However, looking at the differential between the protection for permanent workers and protection for temporary workers, it is higher than in other OECD economies. This may have created incentives for firms to hire more temporary workers, thus creating downward pressure on wages. Also, the labor productivity growth in the services sector is somewhat lower than in the manufacturing sector, though this is common across all economies.

Property income is small due to a strong home/safety bias and low returns from stocks. During the bubble period, the share of property income was above 10% of household income. After the bubble burst, household property income was on a steady decline until the early 2000s. At this time it turned around in line with the sustained recovery, but was still only 4% of household disposable income in 2007, much lower than in other advanced economies, like 20% in the U.S. A key reason behind this is the lower share of risky assets in household balance sheets. The safety bias has not changed much over a 20-year period even though the share of risky assets was rising between 2003 and 2007 when the markets were doing well.

Several factors contributed to the safety bias. First, past financial regulations. Although most of which were removed in the late 1990s, adjustment costs are still high. It may thus take time to shift assets from safe to risky ones. Second, if stock returns are low, which they are, people do not want to hold stocks. Third, historically houses were relatively expensive in Japan. Since houses are risky assets, this may have discouraged households from holding risky financial assets as they may want to optimize their overall portfolio including liquid and illiquid assets. Fourth, some studies show that in Japan, it is not easy to borrow with middle-range interest rates of 5-10%, and those who want to borrow at these rates may be liquidity constrained and need to save cash to finance expected or unexpected expenses. Finally, though it is hard to test whether preferences among Japanese people are different from people in the U.S., some survey results suggest that Japanese people are more risk averse than people in the U.S.

Regarding stock returns, throughout the 2000s, stock dividend yields have been lower than 10-year JGB yields partly reflecting low dividend payouts. Even during the boom period between 2003 and 2007, when corporates were bringing in high profits, dividend payouts were only 2% of GDP. The point is that low stock returns reflect a lower payout ratio. Dividend payout ratios are around 15% now, but that is much lower than in advanced economies, with the G7 average being 50%. The ratio has picked up slightly recently, probably reflecting the unwinding of cross-holding of stocks since the end of the bubble period. The share of foreign stockholders is now 20% and was less than 10% before the bubble period. It is often said that foreigners demand more returns.

The natural question is "where did corporate profits go between 2003 and 2007?" Basically, corporates have been either building up cash or repaying debts to the banking sector. The repayment of debts is helping the JGB market, which is another issue.

The following policy options are quite tentative and deserve comments and feedback.

On wages, it may be helpful to create a regular labor contract with weaker employment protection. The idea is that such a regular contract may encourage firms to hire more regular workers. These workers, who are permanently employed, may have more incentive to accumulate human capital and thus contribute to higher productivity. At the same time, such a measure would address concerns about equity between regular and non-regular workers, which has been a big issue in Japan for years. Finally, this contract would need to safeguard existing contracts for regular workers.

Accelerating labor productivity in the service sector may be the key to boosting wages, since productivity determines wages. One thing that must be highlighted is that labor productivity can be increased either by cutting employment, or creating higher value-added products. This is important to note because cutting employment does not create demand, so corporates should strive to create higher value-added products.

On property income, the bottom line is that with downward wage pressures continuing, distributing more corporate profits to households may help. A distribution issue comes into play with this as stocks are usually owned by the rich. Thus, even though companies may pay out more of their profits, it may be that only the rich will get the benefits. However, if the rich spend more, that may increase employment in the overall economy, thus indirectly helping workers more broadly.

To pay out more, it is important for firms to earn more profits. Productivity can also be enhanced to increase returns, and in this regard, the service sector is key. Tax incentives for holding risky financial assets can be continued. The dividend payout ratio, which has been gradually rising, could naturally rise further, reflecting an unwinding of cross-stockholdings on the part of banks. The issue here is whether public policies may be needed to promote that unwinding process or not.

On the liability side of the household balance sheet, reverse home mortgages have great potential. This idea has been discussed for many years and there has been little progress. However, given the rapid progression of the aging population, this idea still has potential. Some public involvement may be needed because it is very risky for banks to provide reverse home mortgages. One idea would be the provision of insurance by government-affiliated financial agencies.

Second, lack of a market for used housing in Japan has led to banks not being able to provide reverse home mortgages. This needs to be addressed. Expanding consumer credit may also lead to increased demand, but the aggregate impact may not be substantial because those who go to consumer finance companies are relatively low-income households. It is thus doubtful that significant progress could be generated by expanding consumer credit.

It would also be useful to introduce local holidays, which is currently being discussed in the government. Under this system, holidays can be different across regions, so traffic jams may be mitigated thus stimulating tourism. Another idea would be increasing the number of holidays, which may have a stimulative effect on consumption.

Murtaza SYED
There will be three main parts to my remarks on investment First, stylized facts on fixed investment as well as R&D spending in Japan during the boom period preceding the current crisis will be offered. Second, econometric work to understand what has been driving these trends will be presented. Third, the role of public policy in promoting the adjustment in investment that needs to occur in the post-crisis landscape will be explored. In particular it will be argued that export-related investment will need to move away from advanced country demand, which is likely to remain subdued, and be reoriented toward emerging markets, which may require different kinds of products. It will also be argued that investment by SMEs and firms in the services sector will need to be promoted to provide a buffer against external shocks as well as an additional engine of growth.

Part 1. Fixed Investment and R&D Spending in Japan: Recent Trends

Let me begin with some stylized facts on investment and R&D spending in the five-year period of rapid growth in Japan preceding the current crisis. First, notice that much of this investment boom was driven by relatively large firms. The disparity in the annual growth in fixed investment between 2004 and 2007 demonstrates this fact. Second, investment shifted toward the manufacturing sector, whose share in overall investment increased quite dramatically in Japan from around 30% in 2000 to just under 50% in 2007. This is in marked contrast to trends in the U.S., the UK and Germany, where the share of the manufacturing sector in overall investment has fallen over the same period to around 20%. Third, within manufacturing, investment has shifted to the main exporting sectors of Japan's economy, including automobiles, electronics, machinery and steel, whose share in total investment rose from around 19% in 2000 to around 31% in 2006. This has in turn magnified the vulnerability of Japan's economy to external shocks, as was dramatically illustrated during the onset of the current crisis, when both exports and investment contracted significantly.

Fourth, Japan's overall investment to GDP ratio is quite high and is close to the OECD average of 23%, as is the capital intensity of the economy, which has been rising over time. However, the efficiency of capital, or capital productivity, has been on a trend decline. This suggests the need for more innovation to boost the efficiency of the capital stock in Japan. Innovation is directly related to R&D spending, which in Japan is fairly high by OECD standards. However, the R&D intensity of the services sector is very low, and my fifth stylized fact is that the share of the services sector in total R&D spending is the lowest in the OECD. This skewed nature of R&D spending may be contributing to the low service sector productivity observed in Japan in recent years.

Part 2. Econometric Analysis: What Has Been Driving These Trends?

Let me now turn to the second part of my presentation, where I will discuss what has been driving these stylized facts and trends. A standard neoclassical investment model was used to analyze these trends. This model relates the investment rate or the rate of R&D spending to profit expectations through Tobin's Q, as well as several additional variables. The additional variables that were considered include cash flow, which measures the internal funds available to finance R&D spending or investment and is usually used as a proxy for financing constraints in the literature; leverage, which is the amount of debt firms hold in their financing structure relative to equity; and an uncertainty variable motivated by the real options literature, which suggests that when things are more uncertain, firms tend to wait before they invest. The measure of uncertainty is the standard deviation in the returns on weekly stock prices indices of the firms.

An instrumental variables approach was used to estimate the model. The model is in first-differences and includes time dummies to control for firm-and time-specific effects. Under conditions that will be verified through diagnostic tests, this technique yields consistent estimates of the parameters.

When the models are estimated with fixed investment as the dependent variable, the results suggest that firms can be grouped into two baskets: larger firms, manufacturers, exporters and those using capital-intensive technology; and smaller firms, service providers, non-exporters and those utilizing labor-intensive technology. For the first group, the main investment driver is profit expectations, while uncertainty is the main barrier. For this group of firms, we found similar results in Germany, the U.S. and the UK. On the other hand, firms in the second group face significant liquidity constraints in their investment. Despite progress in restructuring, the legacy of excess leverage and dependence on debt financing also seems to hold back investment by the latter, especially high debt firms, suggesting threshold effects. Similar results are found in Germany for this group of firms. However, there is no evidence of significant financing constraints or effects of financing structure on investment in the case of firms in the U.S. and the UK, due perhaps to more diversified financing sources for firms as well as the large amount of liquidity available before the crisis.

Running the same regressions for R&D, we found broadly similar results, though the sensitivity of the explanatory variables other than Tobin's Q are somewhat lower for R&D spending, perhaps reflecting the fact that R&D spending typically tends to have higher sunk costs and higher adjustment costs. For larger firms, manufacturers and exporters, the main investment driver is profit expectations and uncertainty is the main barrier. Smaller firms, service providers and non-exporters again face significant liquidity constraints. The financing structure is less important, although debt financing does hold back R&D spending for firms in the service sector.

Part 3. Policy Implications: Promoting Investment in the Post-Crisis World

This brings me to the final part of my remarks. Based on these results, how can investment and innonvation in Japan be promoted against the backdrop of a changing post-crisis global landscape?

First, raising profit expectations or the return on investment will be key, particularly for firms catering to the export sector. There are a variety of ways that this can be done. One of the obvious ways is for the corporate tax rate to be reduced or having more generous depreciation allowances. Depreciation allowances tend to be especially low for buildings in Japan. The periods for corporate tax-loss carry forwards could also be extended. In Japan, losses can be carried forward for up to seven years, which is relatively short compared to 20 years in the U.S. Finally, there may be greater scope of greater investment and R&D tax credits, or targeted spending in support of research programs as in Germany. This chart shows that R&D tax subsidy rate tends to be relatively low in Japan compared to G7 countries.

Second, lowering uncertainty could also help. This could be done, for instance, by encouraging the use of hedging instruments by exporters. International comparisons suggest that while large firms in Japan do a lot of hedging, they tend to under-hedge credit, commodity and security price risk. SMEs in general tend to not hedge as much, including for exchange rate and interest rate fluctuations. Also, international surveys of investor perception suggest that there is scope to further improve the business climate in Japan in terms of the entry of firms. According to the World Bank, Japan ranks 23rd out of the 26 OECD countries in terms of how easy it is to setup a business. There may also be scope to further streamline labor market regulations as well as bankruptcy procedures. Improving long-term growth prospects by undertaking structural reforms to raise potential growth would also lower uncertainty.

Finally, let me end with some thoughts on boosting investment and R&D spending by SMEs. Our results suggest that financial constraints are particularly severe for small firms in Japan both in terms of R&D spending as well as fixed investments. This can be mitigated by banks broadening collateral to include immovable property, intellectual property rights and receivables. Also, there is scope to encourage venture capital especially for innovative SMEs and high-tech sector firms, for instance by promoting such investments by the GPIF, as is done by public pension systems in a number of other advanced economies. At the same time, reducing excess leverage and promoting restricting would also help SME investment. This could be done by reorienting SME support away from credit guarantees to promote equity investment, as well as further restructuring and exit of non-viable firms. Greater FDI could also promote links with global production networks and technology transfer to SMEs.

Q: Concerning measures with which to promote or enhance private consumption in Japan, especially the present policy of the DPJ, being that they want to distribute cash to households under the child allowance and to farmers under subsidies, the policy implications are that such distributions will enhance private consumption and eventually lead to economic growth and investment. However, it seems that this is not so, since if the government distributes funds as cash allowances, there is no assurance that the recipients will consume those funds, opting rather to save that money to ensure greater income stability. What do you feel about this government policy on the part of the DPJ?

Kiichi TOKUOKA
As a personal view of mine, we understand that the DPJ child allowances are distributed across all households with children. However, looking at micro data, marginal propensity to consume is higher for liquidity constrained or lower-income households. There is, thus, some scope for making child allowances more effective, for example, by setting an income ceiling. However, in order to do this, a way to measure income levels across households must be devised. In that sense, the introduction of a tax ID system should come first.

Q: Officially, from the standpoint of the IMF, do you intend to make a policy recommendation to the Japanese government in terms of such policies regarding enhancement of private consumption through cash deliveries?

Kiichi TOKUOKA
Though I cannot say too much at this stage, we will publish our statement soon. Some recommendations on income transfers will be included in that statement, so that document should be consulted.

Q: Many of the measures you proposed have already been discussed and are impossible to realize. This creates the impression that you might be avoiding addressing some of the most severe questions.

In that regard there are two points. First, the competition in the Japanese economy may be too severe. There are too many companies that are producing similar products, which lowers their profitability and dividends, making their stocks less attractive.

Second, how long can Japan continue to issue JGBs? This may be offset by increasing the consumption or other taxes, but this would in turn reduce consumption. Which would be better?

Murtaza SYED
It is refreshing to hear that we are being too soft rather than too hard; we usually get the opposite at the IMF. In the discussions between ourselves and the government, many of the aforementioned topics are covered. There was a more narrow focus today, but we do share many of the aforementioned points.

Regarding having too many companies, this is what I had in mind when facilitating the exit of non-viable firms was mentioned in the presentation. I share the concern that in some services industries there may be too much competition, which drives down margins and perhaps even incomes. In particular, bankruptcy procedures need to be revisited. Restructuring of some service industries is something that we have been and will continue to advocate.

On JGBs, an important paper was written by my co-presenter Mr Tokuoka in December cautioning that the window is closing on the domestic pool of savings available to finance JGBs, and there may be 5-10 years left.

Kiichi TOKUOKA
Regarding JGBs, Japan cannot continue issuing a huge amount of them, which is clear and shared by the government. Right now, the household sector is not providing as many funds as before. A large amount of funds is now being provided by the corporate sector, which was making huge profits between 2003 and 2007 and de-leveraging. Going forward, the economy will turn around, if it has not started to do so yet. In this case, we think that the corporate sector should start investing again. That point will be a critical juncture because savings from the corporate sector may be less than we see now. Fiscal consolidation is going to be very important and we think that the authorities also have to do a number of things at the same time including limiting expenditure growth and increasing the tax rate. In addition, they will need nominal growth because tax revenues are proportional to nominal growth.

Q: The funds available for venture capital in Japan are quite low. Venture capital is also relatively low in Germany and Italy, and yet the levels of venture capital in the UK, Canada and the U.S. are higher. What is the reason behind these disparities in venture capital rates with Anglo-Saxon countries having higher venture capital levels and other countries having lower venture capital rates?

Murtaza SYED
That is a very good point. There are many similarities between Japan and Germany, including smaller firms tending to have a harder time finding external finance relative to the U.S. and the UK. My hypothesis is that both Japan and Germany are still very much indirectly bank-focused in their financial systems, so that indirect or bank financing is still a very large part of the total financing available to firms.

This may reflect the differences between Anglo-Saxon and other financial systems, but in terms of specifically what is happening in pension funds, one important difference between these two groups is the role that public pension funds play. In the U.S., for example, the California pension fund allocates about 14% of its total assets to alternative investments, which include venture capital and private equity. Similarly, in New Zealand, the pension fund there allocates about 11%, which seems to draw in more interest and confidence from private pension funds as well. On the other hand, in Japan, the government pension fund allocates nothing to alternative investments and Germany has a similar structure.