The Japanese Current Account—Post-Tohoku Earthquake

Date March 17, 2014
Speaker Robert DEKLE(Professor of Economics, University of Southern California)
Panelists for Q&A Discussion Katheryn RUSS(Associate Professor of Economics, University of California, Davis)Eric WEESE(Assistant Professor, Department of Economics, Yale University)
Moderator MIYAGAWA Tsutomu(Faculty Fellow, RIETI/Professor, Faculty of Economics, Gakushuin University)
Materials

Summary

Robert DEKLE's PhotoRobert DEKLE

This research on the current account and the papers which Professors Tsutomu Miyagawa and Kaoru Hosono at Gakushuin University and I are writing have been supported by the grant we have received from the Japan Foundation Center for Global Partnership. I will present detailed graphs and attempt to put the decline in the Japanese current account into a macroeconomic framework.

The current account actually started to deteriorate a few years before the Tohoku earthquake, after the Lehman Brothers crisis. The decline accelerated after the earthquake. Owing to a rapid increase in mineral imports, especially natural gas, we've been saying since 1985 that Japan's current account would deteriorate between 2010-2020. It's a sort of life cycle analysis focused mainly on the aging of the population and the resulting rise in deficits. This has been the standard view for some time.

The monthly Japanese current account fluctuates considerably. My research assistant and I therefore applied a Hodrick-Prescott filter to the current account data. The filtered current account monthly series shows a decline in the Japanese current account starting in early 2007 and then sharply declining in 2009. This sharp decline started before the earthquake. It's interesting that the trade balance became nearly zero for a couple of months or quarters around the time of the Lehman Brothers crisis, recovered, and then turned sharply negative after the earthquake. Services always have been negative at least in the data, but Japan has had large income flows from abroad through its foreign asset ownership, which helped the current account balance remain positive. The main accounting reason for the decline starting in 2007 is that the merchandise trade balance became zero or negative followed by another sharp decline in the current account starting in March 2011.

Again, with regard to the merchandise trade balance, smoothing the numbers through the Hodrick-Prescott filter revealed more medium-run patterns. Total merchandise exports seem to have started to decline in late 2007 while total merchandise imports remained high. The world economy slowed down, exports declined, imports remained high, and the trade balance worsened as a result. Total merchandise imports in Japan then started to fall, which stabilized the trade balance until the earthquake in 2011. Merchandise exports fell immediately after the quake but remained relatively stable, even increasing slightly in 2012. The deterioration in the trade balance was caused by the sharp surge in total imports. Merchandise exports were relatively constant, in fact, they've increased, but imports have increased sharply.

Machinery exports fell immediately after the quake, but have been relatively stable since then. Mineral imports, particularly of liquefied natural gas, sharply increased. Machinery imports also increased somewhat. There was a sharp increase in mineral fuels. Exports remained relatively constant. On the exports side, auto exports fell immediately after the quake but have recovered somewhat. On the trade side, mineral imports were a big factor in the worsening of the current account.

Various goods and services should be examined to analyze the trade balance. For the entire current account, the savings-investment balance should be looked at. Unsurprisingly, the current account worsened after the Lehman Brothers crisis and after the quake. In the famous textbook by Obstfeld and Rogoff, the Great Kanto Earthquake is used as an example due to the very large decline in the Japanese current account after the quake in the 1920s.We think this is because it is very painful for consumers to change their consumption in the short run, so they maintain it. A consumer who views the quake shock as a temporary shock and from which recovery will be swift will not change his/her consumption substantially. However, as the income of the country falls, consumers' savings fall, causing the current account deficit to increase. Assuming this to have been the case after both the Lehman Brothers crisis and the earthquake, the current account deficit worsened twice. It would not be a surprise to see the temporary current account worsening.

The current account surplus by accounting is the difference between total savings (public and private) and total investment (private minus public). There was a large decline in total savings. From a savings-investment perspective, why wasn't the current account deterioration more significant? Private investment also fell in Japan after the Lehman Brothers crisis, as firms discovered that exports weren't growing and that a recession was occurring, causing them to invest less, depressing private investment and helping to ameliorate the decline in the Japanese current account balance. The savings rate in Japan started to recover. Total savings fell after the quake, but there was no commensurate large decline in investment in Japan. Earthquake reconstruction is probably part of this. Because the decline was so localized in Tohoku, overall Japan investment didn't decline much. In fact, there has been a slight increase in investment since 2012, which contributed to the decline in the current account. Savings fell, but the decline was not so great.

What is the role of the exchange rate in the worsening of the Japanese current account? From about 2004 until about 2008, there was a relative weakening of the yen, and then a sharp strengthening in 2009 until fairly recently when the yen in real effective terms depreciated. Short-run fluctuations in both the nominal exchange rate and the real exchange rate are driven by many factors, but, in the very long run, we'd like to think that exchange rates are driven by real fundamentals (productivity differences between countries—the Balassa-Samuelson Effect). The yen was greatly inflated between 2009 and 2013, and the yen-dollar rate is approximately in equilibrium today. The exchange rate fluctuates greatly in the short run which could affect growth very negatively, output and trade balances in the short run, and this equilibrium model allows you to evaluate this. Today's rate seems close to where it should be. Between 2003 and 2007, the yen was possibly a little too weak. It then started to appreciate until it was clearly greatly inflated.

The current account trade balance and the yen equilibrium real exchange rate are not completely divorced. If they were incorporated into a model, the appropriate exchange rate would be fairly close to what it is today. In the short run, the exchange rate could overshoot wildly, causing various detrimental effects on various economies.

The Japanese nuclear power plant shutdown caused a need for energy supplies for the manufacturing sector and a rapid increase in mineral imports, which resulted in a deterioration in the trade balance. The quake also caused a temporary shock to Japanese productivity, but there was no reason for consumption to decline so much, and the need for reconstruction caused investment to rise as savings declined resulting in a decline in the current account. Can these two stories be integrated? Nobu Kiyotaki, Hyeok Jeong, and I have written a paper that tries to relate shocks to particular commodities caused by the earthquake and other factors to changes in aggregate inflows and the current account through an intertemporal macroeconomic trade model. We generally find that where a productivity or quake shock damages the productivity of some firms, those firms will stop exporting and the real exchange rate appreciates. Current account deterioration then will occur. However, this is preliminary research.

Where do I think the Japanese current account is headed? The projection is that Japanese household savings will continue to fall. Corporate savings have sustained total savings. Government savings may remain flat and the current account will continue to decline over the medium run. I think the economics show that and many models have shown that. The debt situation is somewhat worrisome when you have large current accounts.

However, Abenomics could change the current account to positive if permanent structural changes in Japan generate a real turnaround after all of the public and private investment, resulting in an increase in the permanent growth from approximately 1% to 2.5% in real gross domestic product (GDP) terms. Then there could be a large increase in GDP with consumption spread out, causing savings to recover somewhat and helping to improve Japan's current account balance in the future from a macroeconomic perspective. Thank you very much.

Participants' Opinions

  • Eric WEESE: I found the discussion about permanent shocks versus temporary shocks to the economy to be interesting. Normally, when something like a quake destroys capital, it is thought of as a temporary shock, but it can also be argued that there's been a permanent technology shock—the price of nuclear power either is infinite or it increased a lot, so in a certain sense, in this particular earthquake, there's the potential for both a temporary shock and a permanent shock, so maybe the data can be seen and we can ask, "Given the data, does this look more like a temporary shock or a permanent shock?" There was also a great graph about the Japan-U.S. real exchange rate which showed a large overvaluation in the middle. It seems that was a time when there was a lot of argument, at least in North America, over the Japanese economy being configured unfairly for export promotion. A similar argument has appeared recently with regard to the Chinese. The model is saying that it's overvalued during a period when there was argument about the Japanese taking unfair advantage of some foreign trading partners.
  • Katheryn RUSS: I'm curious in terms of the long-term implications for the current account. Increasing growth might reduce the current account, but at the same time given what's going on in Eastern Europe and Russia right now, I'm wondering if fossil fuel prices may go up and if that could impact growth. To the degree that it would reduce growth, maybe it would also improve the current account, but to the degree that it increases the prices of imported fuel, maybe it would worsen the current account, so I'm curious about what the net effect might be.

Speaker's Response

Robert DEKLE
You're saying that shutting down the nuclear power plants is a temporary, not permanent, shock. I think it's a political problem. If it's a permanent negative shock, you would have consumption declining greatly, so the current account would not worsen so much, but the fact that it worsens suggests that people see the quake as a temporary shock. The energy issue could be more of a permanent shock. It's been continuing for much longer, which should result in seeing a slight improvement in the current account. A negative terms of trade shock may not deteriorate the current account as much, but that's worse than a temporary terms of trade shock because real GDP growth will fall with a permanent terms of trade decline, so if one cares about long-run real GDP growth, then a deterioration in the current account is not so bad.

Q&A

Q1: If we assume that we are going to see a very sharp deficit of current account situations and that it's bad for the Japanese economy, what should policymakers do?

Robert DEKLE:
I don't think it's bad at all for the economy to have a current account deficit. But what should you do to reverse the current; it's very hard to change fertility and household savings patterns. Investment for future growth is important. I'm not against more energy independence to reduce. This rapid increase in mineral imports is worrisome, and it cannot continue indefinitely, so helping decrease that is important. I think that's one thing the government could do.

Moderator MIYAGAWA Tsutomu:
An increase in consumption may lead to an increase in imports. But from the welfare point of view, the deficit in the current account is not bad. However, from the viewpoint of Abenomics, the increase in consumption leads to domestic investment and expansion of the domestic production capacity, so this current movement in the current account is different from the scenario created by Abenomics. So, I'd like to focus on the domestic investment movement. I think that we can explain the current movement of the current account or trade balance by the hysteresis effect which means that due to the fixed cost of domestic investment, Japanese firms are reluctant to make investment in Japan. I'd like to know your opinion on this type of argument.

Robert DEKLE:
So, you're saying investment is artificially low and that firms don't want to invest in Japan because they think the marginal productivity is not so high?

Moderator MIYAGAWA Tsutomu:
Yes, due to the fixed cost, although the exchange rate has returned to the pre-Lehman Brothers collapse level, the GDP ratio has not been recovered.

Robert DEKLE:
So, firms have permanently left Japan, and they're not going to return. The investment is going to be permanently low. That's not a good outcome. There's scope for policy to intervene to try to increase investment back to where it should be.

Moderator MIYAGAWA Tsutomu:
One of the policy options is to decrease the fixed cost in domestic investment.

Robert DEKLE:
I completely support that notion of reducing fixed cost. Lower corporate taxes are a great thing as well as forgiveness of past debt; all those policies should be implemented. Fixed costs have been high. The prolonged strong yen may have made it difficult for firms to return to raise the investment level and that suggests a scope for policy.

Moderator MIYAGAWA Tsutomu:
The other policy option is that the government promotes the entrance of high-productivity firms to the market. If such firms enter and grow, the current trend in the current account from the savings-investment approach will change.

Q2: Could you explain in more detail how to calculate the equilibrium exchange rate?

Robert DEKLE:
This is just a model based on the Balassa-Samuelson Effect. There are two kinds of goods in the world. One is goods traded freely between countries which are called traded or imported goods. And then there are other goods such as haircuts. In the trade balance, some of the services are traded such as consulting, financial, and legal, but many other services are very difficult to trade because the cost is so high, such as restaurants and bars.

The Balassa-Samuelson Effect states that if you have different productivity growth in the goods that are traded, it doesn't make a difference to the prices because trade makes the prices very close in the two countries. For non-traded goods, if you have more rapid productivity growth in non-traded goods in Japan compared to the United States, for example, then you have faster wage growth in Japan and demand there increases more, and there is more demand for services so the service prices will be bid up relative to the service prices in the United States. If you take the price ratios in the two countries which include both traded goods and non-traded goods, the prices of services are going to increase faster than that of non-services. The country with the more rapid productivity growth will have a more rapid appreciation of the exchange rate.

Q3: From the saving-investment balance point of view, many are saying that the current account in Japan is deteriorating and will deteriorate further. How do you explain your focus on the deterioration of the current account balance from the viewpoint of trade balance and current transfers? From this point of view, there are some who are saying that trade balance deficit would be offset by current transfers or surplus.

Robert DEKLE:
Consumption and investment will remain high so the merchandise trade deficit will become increasingly negative, and there might be an improvement in fuels or mineral fuels, but there could be more expansion in services. And once it turns negative, there's a problem in that you start eating into the capital, and then the income size will fall. I understand the concern that a current account with large fiscal deficits can be risky in the future, if market sentiment turns against holding this. My instinct is that production in Japan is going to recover, creating a positive shock to output or productivity growth that will help the current account. Negative current account balances are not bad for Japan, but there is a concern given the high deficits of some risk in the future. To relate these components to the current account, it would be the case that merchandise imports and service imports will continue to rise—a story consistent with a deteriorating current savings and investment balance. I've never heard a good explanation relating the trade balance components and the current account. That's a big gap in our knowledge.

Q4: I have two questions. In the Q&A session, you said that the current account deficits are not problematic and your PowerPoint materials state, "Would not want to have large government debt and large current account deficits at the same time." So this means that your view is that current account deficits held by large governments are problematic? The second question is: based on your prediction, Japan will have large current account deficits from now on. Some economists in Japan are worried about the country's current account deficits because it means that they must be financed by foreign capital inflows, then the foreign capital inflows might be very unstable, and Japan will not be able to maintain long-term interest rates. If interest rates become higher, that might create some obstacles for investment. Do you think that the current account deficits might increase domestic interest rates and that they might cause problems?

Robert DEKLE:
The questions are related. If you have large government debt, there will be a need to roll over or continue to increase Japanese government bonds (JGBs) and someone has to buy them, and if one starts to run into current account deficits—right now, the Bank of Japan (BOJ) is buying marginally quite a bit—it's true that if you have to start to rely increasingly on foreigners, then the interest rate is going to increase, raising debt, which would make the situation worse.

If you have large government debt and a large current account deficit at the same time, it's problematic. There could be tax policies, but raising savings is not easy. I don't know if low investment is desirable, either. High private investment is desirable to increase growth in the economy, but it does raise risks. Is it worth sacrificing the risk and instability for lower growth? There's literature that high growth is riskier than low growth. A current account reversal in many countries can be accomplished by lowering consumption and lowering investment, but is that what we should do in Japan now? I don't know.

My coauthor Nobuhiro Kiyotaki's view on exchange rates, with which I agree, is that it is still not completely independent of the trade balance and current account deficit and savings and investment. It looks like they're completely independent because the data can't precisely estimate the relationship between the fundamentals and the exchange rate.

Katheryn RUSS:
In terms of elaborating on the problems with the data that make it difficult to link the trade balance directly with the exchange rate, there are some nice studies by U.S. economists showing that there's a lot of double counting in the trade data. When China exports an iPhone to the United States, the full value of the iPhone is counted in Chinese exports whereas Japanese imports and U.S. design are a part of that, so the value added is only a few dollars but the exports look like several hundred dollars for that iPhone. When we think about changes in real activity over time, we know that the Japanese manufacturing sector did hollow out to some degree during that period of the endaka. The puzzle we face is: how do we relate the trade balance data to what we observe in the exchange rates?

Q5: It is interesting that the pattern of changes in the real effective exchange rate and the changes in the current account balance of payments of Japan do not correspond with each other. I think that the decline in the real effective exchange rate is due to so-called internal devaluation. May I just clarify the point you made with respect to calculation of the real effective exchange rate. Did you say that in calculating it, you used the so-called general prices indices?

Robert DEKLE:
Yes, that's right.

Q6: The current account balance of payment began to deteriorate around 2005 to 2007 due to the decline in the competitiveness of Japanese manufacturing goods. In Japan, service prices have been declining relative to other countries, whereas the prices of manufacturing goods might in relative terms have been increasing vis-à-vis such countries as China and Korea. Therefore, in terms of competitiveness, Japan lost vis-à-vis emerging countries. That's the reason why the current account balance of payment of Japan began to deteriorate around 2005-2006 despite the fact that the real effective exchange rate has been declining continuously. Is that a correct understanding of what you have said?

Robert DEKLE:
I haven't really looked at Japan in terms of its Asian trade partners. Of course, recently, Japanese trade with China has been much more important in terms of magnitude. There is entire literature on whether the price indices accurately measure the actual prices faced by importers and exporters. Even in a multilateral sense, the yen is allegedly weaker today than it was in 2012.

*This summary was compiled by RIETI Editorial staff.