Last Thursday night, I dashed off a post in response to accusations being made by Chinese and South Korean critics of Abenomics that Japan is now engaging in currency manipulation. When I started writing, I thought that I was going to dismiss such accusations, because Prime Minister Abe has made an increased inflation target an explicit goal of his monetary policy, and instructed the newly installed Governor of the Bank of Japan to meet that target. However, despite the 25% depreciation of the yen against the dollar since it became clear last fall that Mr. Abe, running on a platform of monetary stimulation, would be elected Prime Minister, prices in Japan have not risen.
It was also disturbing that there were news reports last week that some members of the Board of Governors of the Bank of Japan voiced doubts that the 2% inflation target would be met.
Some of the members of the Bank of Japan (BOJ) board were doubtful about achieving the 2% inflation target projected by the bank within the two-year time frame, according to the latest minutes of the policy meeting.
Why a 25% decline in the value of the yen in six months would not be enough to raise the rate of inflation to at least 2% is not immediately obvious to me. In 1933 when FDR devalued the dollar by 40%, the producer price index quickly jumped 10-15% in three months.
Moreover, the practice of currency manipulation, i.e., maintaining an undervalued exchange rate while operating a tight monetary policy to induce a chronic current-account surplus and a rapid buildup of foreign-exchange reserves, was a key element of the Japanese growth strategy in the 1950s and 1960s, later copied by South Korea and Taiwan and the other Asian Tigers, before being perfected by China over the past decade. So despite wanting to defend the new Japanese monetary policy as a model for the rest of the world, I couldn’t conclude, admittedly based on pretty incomplete information, that Japan had not reverted back to its old currency-manipulating habits.
My expression of agnosticism invited some pushback from Scott Sumner who quickly fired off a comment saying:
I don’t follow this. Why aren’t you looking at the Japanese CA balance?
To which I responded:
Scott, Answer 1, CA depends on many things; FX reserves depends on what the CB wants. Answer 2, I’m lazy. Answer 3, also sleep deprived.
Well, I’m sticking with answer 1, but as I am somewhat less sleep deprived than I was last Thursday, I will just add this tidbit from Bloomberg.com
Japan‘s current-account surplus rose in March to the highest level in a year as a depreciating yen boosted repatriated earnings and brightened the outlook for the nation’s exports.
The excess in the widest measure of trade was 1.25 trillion yen ($12.4 billion), the Ministry of Finance said in Tokyo today. That exceeded the 1.22 trillion yen median estimate of 23 economists surveyed by Bloomberg News.
Prime Minister Shinzo Abe’s revamp of Japan’s central bank to focus on ending deflation paid off when the yen today slid past 101 for the first time since 2009, helping exporters such as Toyota Motor Corp. (7203), which now sees its highest annual profit in six years. Sustaining a current-account surplus may help to maintain confidence in the nation’s finances as Abe wrestles with a debt burden more than twice the size of the economy.
“The currency’s depreciation is buoying Japan’s income from overseas investment at a pretty solid pace,” said Long Hanhua Wang, an economist at Royal Bank of Scotland Group Plc in Tokyo. “A weaker yen provides support for Japanese exports.”
The cost of a weaker yen is higher import costs, reflected in a ninth straight trade deficit in March. The current-account surplus was 4 percent lower than the same month last year and the income surplus widened to 1.7 trillion yen, the highest level since March 2010, the ministry said.
So contrary to what one would expect if the depreciation of the yen were the result of an inflationary monetary policy causing increased domestic spending, thereby increasing imports and reducing exports, Japan’s current account surplus is approaching its highest level in a year.
Then, on his blog, responding to a commenter who indicated that he was worried by my suggestion that Japan might be engaging in currency manipulation, Scott made the following comment.
Travis, I had trouble following David’s post. What exactly is he worried about? I don’t think the Japanese are manipulating their currency, but so what if they were?
OK, Scott, here is what I am worried about. The reason that currency debasement is a good and virtuous and praiseworthy thing to do in a depression is that by debasing your currency you cause private economic agents to increase their spending. But under currency manipulation, the desirable depreciation of the exchange rate is counteracted by tight monetary policy designed to curtail, not to increase, spending, the point of currency manipulation being to divert spending by domestic and foreign consumers from the rest of the world to the tradable-goods producers of the currency-manipulating country. Unlike straightforward currency debasement, currency manipulation involves no aggregate change in spending, but shifts spending from the rest of the world to the currency manipulator. I don’t think that that is a good thing. And if that is what Japan is doing – I am not saying, based on one month’s worth of data, that they are, but I am afradi that they may be reverting to their old habits – then I think you should be worried as well.
Uneasy Money 
I’ve a vague impression that Scott Sumner has a blind spot with regard to exchange-rate manipulation. Despite being a sticky-price man he’s reluctant to accept that it’s possible to rig the real exchange rate to any significant extent.
LikeLike
David, but didn´t Japan have a trade balance deficit, the spending part of the CA?
LikeLike
May I ask a clarifying question?
If japan wanted to lower the value of the yen while maintaining tight monetary policy then how would it do this ? My (probably simplistic) view sees them devaluing by printing up yen and buying dollars with them. However if they try to counteract the domestic effects of this with tighter monetary policy won’t this be done by reducing the yen money supply and increasing its exchange value again ?
Can you explain how “exchange-rate manipulation” works ?
LikeLike
Rob, I think you are Right. Only of there si some capital control it is possible to debase the yen and simultanously to get an internal monetary contraction.
LikeLike
I suspect that there is a lag with the CA balance – a depreciating Yen would impact exports before monetary policy would impact imports, due to structural factors like supply chains and embedded expectations (i.e., propensity to save).
I believe that financial repression is the typical way to rig the CA balance – interest rate controls and forced savings, and (for non-convertible currencies) capital controls. While I am not a Japan expert, my understanding is that Japan has an extensive system of forced savings and financial system control (but it also has a floating fx, which should free up monetary policy), which forces up the savings rate, thereby increasing the CA surplus. Not sure how much that sterilizes monetary stimulus, though, as that would probably depend on CB reserves.
LikeLike
Kevin, I do believe that governments can manipulate the real exchange rate to some extent. But I don’t believe that hurts other countries. The Fed determines nominal spending in the US, not Japan.
I also don’t understand David’s sterilization argument. The BOJ just announced a huge QE program. And the yen’s been falling on rumors of monetary stimulus.
LikeLike
David, to confirm your suspicions of potential Bank of Japan sterilization, what would you expect to see occurring on the BoJ’s balance sheet?
LikeLike
Kevin, Well, he seems to think that we should be indifferent to whether they practice currency manipulation, because they are subsidizing our consumption of their goods. That argument has a certain validity, but in the context of a business cycle downturn, and perhaps in other situations, I think that it is a dangerous policy.
Marcus, Well, I didn’t know that, but it seems that you are right. That does blur the picture somewhat. But I have not claimed that Japan is in fact engaging in currency manipulation, as it clearly has done in the past, just that there are some signs that it may be doing so again. I will be happy if my suspicions are shown to be unfounded.
Rob, Good question. The idea of currency manipulation (or exchange rate protection as Max Corden referred to it in his classic paper) is that the monetary authority intervenes directly in the foreign exchange market and sells its currency on the market driving down the value. By sterilizing the inflow of foreign exchange, the monetary authority maintains an excess demand for cash which forces the public to restrict spending in order to build up their cash balances, producing an balance of trade surplus. If you imagine that the monetary authority pegs its exchange rate, the policy is clearly feasible despite the tendency of the policy to cause an appreciation of the currency, provided that the monetary authority is willing to accumulate foreign exchange reserves to any extent necessary to support the policy. Another way of supporting the policy is to impose a reserve requirement in terms of foreign exchange on the domestic creation of currency and deposits.
Luis, I don’t think formal exchange controls are actually necessary, but they certainly can help to facilitate the policy.
Dan, You are right, the immediate effect of a depreciating exchange rate is usually to increase the balance of trade deficit. It is only later that the shift in foreign trade patterns dominates and begins to cause a reduction in the trade deficit.
Scott, Yes, in principle, the Fed can always compensate for any reduction in aggregate demand that results from exchange rate protection by foreign countries. That doesn’t absolve them for responsibility for imposing a reduction in aggregate demand on the rest of the world.
Yes BOJ announced a QE program. The question is whether they are doing what they said or whether they are reverting to old (bad) habits. The yen has been falling in the expectation that the yen would fall, that fact, in isolation, doesn’t allow us to distinguish between the hypothesis that the proximate cause of the depreciation is QE or exchange-rate protection.
JP, Well, we need to know the mechanism by which BOJ would be sterilizing, capital controls, reserve requirements, or open market sales. However, under almost any conceivable approach, the foreign asset holdings of the BOJ should be increasing if they are in fact engaging in currency manipulation.
LikeLike
I still say people, including my favorite bloggers, are missing the bigger picture.
1. Japan (the BoJ) tried QE from 2001-6, and John Taylor raved that it was a success. It coincided with Japan’s longest postwar expansion, though not particularly robust. Then they stopped QE, and went right back to ZLB-perma-gloom. So four years of QE was not enough.
2. The snivelers are already whining that QE should be stopped in the USA—but if four years was not enough in Japan, what makes anyone think less than four years is enough in the USA? Inflation just came in at 0.7 percent, y-o-y, PCE. We might need six years of QE, or….
3. We might need permanent QE. Why have global sovereign debt yields fallen for 30 years running, until now everyone is converging in ZLB land? Sumner days he does not know, and I suspect global savings gluts. People have to save money regardless of interest rates, and more people than ever can save (not spend every penny to survive). I have to save for retirement, and if we are in ZLB I still have to save….
4. Which leads to problem #4: Okay we sensibly go to perma-QE, and you get an escalating Fed balance sheet, and huge revenues for the Treasury Department. I like this actually. But, under the dubious system of accounting in place, the Fed, when it sells its hoard of QE bonds, is supposed to record a capital loss or gain, and either forward gains to Treasury, and seek compensation from Treasury.
This is nutty! Nutty! The Fed printed (digitized) money and bought bonds. When it sells the QE bond-booty it should post a huge capital gain, and forward that money to the Treasury.
If I successfully counterfeit $10 million, and buy $10 million in US bonds, and bank them, and then sell five years for $9 million (a “loss” due to a rise in interest rates) I do not have a “loss” of $1 million in real life, I have a capital gain of $9 million. I have $9 million in the bank, and I started out with nothing.
I wold like Glasner’s observations on the proper accounting treatment, after sale of the Fed-QE hoard.
LikeLike
David, I don’t agree that a Japanese weak yen policy imposes lower AD on the rest of the world. US AD is determined by the Fed, not the Japanese.
On the other hand a weak yen also does not subsidize the US in any way that I can see.
I’m pretty confident that the BOJ will carry through with its QE policy, so if that’s the criterion, I don’t think we need to worry about currency manipulation in this case.
I would add that they don’t need to buy foreign assets to manipulate the value of their currency—any form of government saving will do the job, if not offset by private dissaving.
LikeLike
@Benjamin Cole,
I’m not clear on the last part of your post:
“4. Which leads to problem #4: …
…
…I have $9 million in the bank, and I started out with nothing. ”
Here’s how I see it: After Tsy deficit spends $10M and the Fed buys $10M of Tsy debt from non-banks (assuming they purchased the debt from Tsy), and everybody pays their loans off (assuming everyone started w/ an all zero balance sheet: i.e. non-banks pay back bank loans and banks pay back reserve loans), we have:
Tsy:
Assets: $0
Liabilities: $10M Tsy debt
Negative Equity: $10M
Fed:
Assets: $10M Tsy debt
Liabilities: $10M reserve deposits
Equity: $0
Banks:
Assets: $10M reserves
Liabilities: $10M deposits
Equity: $0
Non-Banks:
Assets: $10M deposits
Liabilities: $0
Equity: $10M
Now the Tsy debt loses $1M of value:
Fed:
Assets: $9M Tsy debt
Liabilities: $10M reserve deposits
Negative Equity: $1M
And the Fed sells its Tsy debt back to non-banks:
Tsy:
Assets: $0
Liabilities: $10M Tsy debt
Negative Equity: $10M
Fed:
Assets: $0
Liabilities: $1M reserve deposits
Negative Equity: $1M
Banks:
Assets: $1M reserves
Liabilities: $1M deposits
Equity: $0
Non-Banks:
Assets: $1M deposits, $9M Tsy Debt
Liabilities: $0
Equity: $10M
When the Fed sells its Tsy Debt, it simply erases $9M of reserve-liabilities on its books. It doesn’t make a profit from that. In fact it’s left holding the bag with $1M in remaining liabilities. Now the Combined Tsy/Fed balance sheet went from $10M negative equity to $11M negative equity. This is just normal double entry accounting. Nothing unusual here. The Fed didn’t counterfeit anything, it wrote out its own IOUs (Fed deposits) for Tsy IOUs (Tsy debt). The usual exchange of IOUs that is the trading of financial assets. When a party gets its IOUs back, it erases them.
LikeLike
Benjamin, I agree with most of what you say, but I am not ready or willing to get into a discussion of accounting rules with you (or anyone else, for that matter). Sorry.
Scott, If (underline “if”) the Japanese weak yen policy is associated with a tight monetary policy that is sterilizing (to use an imprecise term, for want of a better one) the inflow of foreign exchange that would normally follow from a depreciating yen, then Japan is indeed reducing aggregate demand for the rest of the world. Aggregate demand in the US is determined by many factors, of which Fed policy is the most important, but not the only, component.
A weak yen means that we get to buy more stuff from Japan. In that sense it’s good for us, but that is a long-run view that ignores the very important transitional adjustment costs associated with more imports from Japan that compete with domestic products when unemployment is already high.
I have great confidence in your judgment about what the Japanese will do, but that still doesn’t mean that I don’t worry that even you might be mistaken.
Yes perhaps other forms of government saving could accomplish roughly the same objective, but the question is what is the BOJ actually trying to do, not whether someone else could do the same thing.
Tom, Thanks for the accounting lesson, but I am not getting involved. You and Benjamin are on your own.
LikeLike
Tom Brown:
I think I savvy what you say, though it has been 30 years since I took basic accounting courses.
Let’s look at it this way, and see if I am crazy:
Okay, a central bank literally prints $10 million. Cash, greenbacks.
They buy $10 million in national government IOUs, bought from private investors. The private sector now has $10 million in cash—they have been paid in full.
The national government now pays interest on those IOUs to the central bank, and the central bank just gives the money back to the national government.
Eventually, the central bank sells the national IOUs to the private sector again, and gets back $9 million.
Under current accounting rules, the central bank must now seek $1 million from the national government.
I like it better that the central bank transfers the $9 million in profit (a kind of seigniorage) to the national government.
Related question: What if the central bank just holds the bonds to maturity? Do they still report a “loss.”
LikeLike
@Benjamin Cole,
First let me warn you that I’m no accountant (or expert of any kind on this stuff)! I just find thinking in terms of simplified balance sheets helpful sometimes.
Regarding your greenback scenario, you are correct: accounting rules say the Fed is in the hole $1M. Greenbacks are treated no differently than electronic deposits on the Fed balance sheet (they are liabilities to the Fed after being sold to banks). Tsy mints the paper money, and the Fed pays the production costs (not face value!) to acquire them. While at the Fed they are essentially worthless. Banks pay face value for them. Coins are different: they are purchased by the Fed at face value from Treasury (Tsy absorbs the production costs… and actually loses money on pennies and nickles I understand!), and thus coins are assets while at the Fed (and erased as assets when sold to the banks). But it still doesn’t affect your example: the Fed ends up in the hole by $1M whether buying (or selling) the Tsy debt with coins, paper money, or electronic deposits
That’s my understanding anyway!
If the Fed were to remit the $9M back to Tsy, that would be a form of government self funding. Tsy seigniorage on coins is too I guess, but there’s a limit to how much of that goes on.
Now regarding your related question:
“Related question: What if the central bank just holds the bonds to maturity? Do they still report a “loss.””
No, in that case I don’t think so because the Tsy still must pay back the principal at maturity even though the Fed holds them and that principal does NOT get remitted to Tsy*. Of course at maturity the Fed must get paid the full $10M back.
*My source for that was JKH who writes articles at monetaryrealism.com.
Actually, if the Fed were just made a desk at Tsy (i.e. told to take direction from Tsy), and was allowed/instructed to buy debt directly from Tsy, then this would be a form of self funding too, since even if the principal were not remitted, no private sector funding would be required at all for Tsy. It may not seem like much of a difference (since the Fed buys so much Tsy debt), but I think it matters that Tsy must turn to the private sector (not the Fed) to fund the Treasury General Account (TGA), and that it can’t overdraft the TGA, and that the the Fed has a large measure of independence.
LikeLike
BTW, not to say there’s anything wrong w/ gov self funding: but that’s just a very different system than what we’ve got now. I’m agnostic as to whether self funding is a good idea.
LikeLike
Self funding is essentially the trillion dollar coin idea (some serious seigniorage!) which nobody dared to do, but which I guess technically remains a legal possibility.
LikeLike
another article another adjective!
I may not be a fan of Hawtrey but I certainly respect him
LikeLike
Tom Brown–
I enjoyed your commentary.
Yes, monetizing the debt, or seignorage is the way to go now.
I am not disputing what current accounting rules are. I am disputing that they are a good idea.
At this point, I think the Fed should print (digitize) money and put it into circulation. The fact that the market thinks they have to buy back the bonds is depressing the economy, and leads to a strange “loss” that has to be made up by taxpayers.
LikeLike