What About Japan?

- All right, well, we are delighted to have Han Lustig from right here at Stanford talking with a great title. What about Japan? And so take it away hanno. So, so do you want, sometimes people talk, speak uninterrupted for 45 minutes and then the other, but you're an economist, so can we interrupt all the time,

- Please?

- Okay,

- Good. Yes. I think that's gonna, that's gonna work better. Okay. Well, thanks so much for giving me this opportunity. Really excited to get to talk to you about this paper we have. What about Japan? This is joint work with Ly Chen, who's at the St. Louis Fed and Hal Cole at upen. And, and in a way, this is actually A-U-C-L-A product because we, we were all at UCLA when we met each other. LY was a student of Hal and mine. And you mentioned the title. So let, let me actually tell you how we ended up with this title. When I started working on fiscal sustainability, probably six or seven years ago when I started giving seminars, there would always be one person in the seminar room you could set your clock to it, who would raise his or her hand and say, wow, what about Japan? And how do, I've been getting that question every year since about 2005.

- You must have

- Been getting that question

- Right. Every fiscal theory seminar, someone thinks this is a brilliant new observation.

- Yes. And, and it was a bit frustrating because I didn't really have a good answer because I didn't know anything about Japan. And so then I decided to get with Hal and Neely to actually dig into what's going on in Japan and, and try to make sense of, have you lived in Japan? I have been to Japan once. In fact, I went there to give this paper, and I can tell you later what, what they told me about the paper. I got some feedback on it. But, so that's essentially the, the sort of, the genesis of the paper is trying to address that question because whenever you talk about fiscal sustainability, people like to come up with sort of a counter example suggesting that maybe we shouldn't be too worried about the government center temple budget constraint. And I, and, and so what we realized from looking at Japan is that it's certainly not the right lesson. Okay. So what I wanna do first is, is sort of talk a little bit about that dynamics First for just the central government in Japan. I'm gonna try and convince you that that's not the right way of looking at it. And then I'm gonna think about the entire public sector where we consolidate the central government with the central bank and pension funds and publicly owned financial institutions. And then we're gonna actually make progress on what's going on. Just to set the stage for the discussion, let me remind you why Japan has been running large deficits. This is probably not news for anyone here, but Japan is at the leading edge of this demographic transition that most advanced economies are undergoing. As you know, fertility rates have dropped tremendously around the world, whereas at the same time, life expectancy is going up. And that puts a lot of pressure on the budgets of governments, especially in countries where there's a pay as you go or unfunded pension system. And so over the past 20 years, Japan social security payments have increased from 17 to 35% of the government's budget. So that's the main source of these deficits. The Ja, the Japanese government has made generous promises to older Japanese people and they're struggling to keep those promises. So as a result of that, Japan's government is running large deficits, primary deficits of around 5% per annum between 1997 and 2023. I'll show you that in a second. And as a result of that, it has one of the highest at GDP ratios in the world, but so far without a fiscal crisis. And so then the question is how, why, and I'll start off by saying the answer is not that Japan is sort of applying some of the R lower than G magic. It's not in that sort of Goldilocks region of, of the parameter space. So that is not what's going on. Why is it interesting for all of us to study Japan? Well, I think it's interesting because of what I said, they're at the leading edge of this demographic transition. And, and it looks like certainly other advanced economies are gonna deal with or already dealing with similar challenges, certainly in, in Europe, US is a little bit behind in terms of the demographics, but, but it looks like we might be on a similar trajectory. So studying Japan is, is kind of interesting because in a way they've been, they've been running an experiment for us, and I think it's kind of important to try and see if we can understand what happened there. So,

- Hannah, can I ask

- Yes.

- How much of the debt is due to they're trying to use Keynesian stimulus to get the economy going? Do you know offhand?

- I don't have a, an exact number for that. That's, that's actually, that's a good question. I I should know. I Yeah, but I think

- I, I'm sure sure that that's the minority, but still, I'm just wondering Yeah, if it's still noticeable even.

- Yeah, my impression is that the, the biggest contributor to the secular increase in deficits, just the, the pensions, but that is

- The usual story told, is that they went on this big, you know, trains to nowhere where fiscal expansion. So I'd be curious to see no social security.

- Yeah, because I calculated that 27% of the US debt is from all of the stimulus packages during the financial crisis and COVID plus accumulated interest.

- I see. Right. So I suspect it, it's a bit smaller in Japan, but I I should have a, a more precise answer. That's a good question. So what have they been doing? Let me sort of give you the bottom line and then, and then provide some more detail just in case I, I don't get to it later on. What they've been doing, if, and this is what you notice when you consolidate the central government with the central bank, the Bank of Japan and all the pension funds, et cetera. And you look at the consolidated balance sheet is to realize balance. They've been essentially running a risky carry trade and by taking risks in various markets, currency markets, equity markets, bond markets, they're earning an additional two to 3% of GDP per annum on average. How do they do that? Well, first there's risky maturity transformation to the tune of one GDP. How do they do that? Well, they, they borrow at floating rates through the Bank of Japan, which issues reserves. I'll come back to how that works, but that's what happens when the Bank of Japan buys back government bonds and issues, reserves. You're borrowing in floating rates, and then they turn around and they invest mainly through the pension funds in risky assets, including longer maturity bonds abroad in stocks. And they're also in the process, as I said, investing half of Japanese GDP abroad and without hedging currency risk. So what does that mean? Well, they're doing a classic currency carry trade. Japan is, or the Japanese yen is a funding currency. If you're a carry trader, suppose you're at a hedge fund and you're running a currency carry trade, what you would do is just that you would borrow in yen and invest in US dollar or other foreign currencies. The Japanese government is doing just that to the tune of one half of GDP. So if you add all of these different risk premia, they're harvesting in different pockets of the market. If you add it all up, it adds up to two to 3% of GDP, which by the way, that's a big deal. That's actually a large number. Now of course, they're taking risk, right? This is not a free lunch. As a result of this, there's a huge duration mismatch on the government's balance sheet that I'll talk about when rates go down. That's great news because that creates a lot of extra fiscal capacity for the Japanese public sector. Why is that? Well, because their liabilities are bias towards floating rate because of what the Bank of Japan's doing. They have a lot of assets that are longer duration. And so actually the way the duration of their net liabilities is, is negative actually, but it's surpluses are very far in the distant future. So if rates were to go up dramatically, and it seems like that's starting to happen, that would be very bad news for the fiscal capacity.

- Yeah. Question please. Some years ago there was, something went on in Japan, and I read the monetary policy committee, which they're all translated into English. Have you done a kind of search through all those things to see whether or not they say what you're talking about to each other?

- Ah, that's a great question. So the language that is used amongst central bankers and, and that applies to the Bank of Japan as well, rarely makes explicit references to fiscal challenges for obvious reasons. But I'll try to convince you that quantitative easing, the way it's been carried out in Japan was largely an exercise in financial repression, was just a way of the government to try and lower its cost of funding. And one little piece of evidence that I'll mention right now is if you thought it was all about the zero lower bound, then the question is why is the Bank of Japan still buying bonds today? Bonds, even though inflation is actually running at 3%, they're pretty, wow. They're, they've solved a zero lower bound problem. They're a long ways away from it, but they're still buying bonds. That of course, suggests that maybe what really doing was, was, was, was actually financial repression along the lines of what the Fed and the treasury were doing in the second World War, for example.

- So I, I think, yeah, my, my intuition would've been that the long duration liabilities, you know, that, that, that part of it, you know, rates go down that Yeah. Gets more expensive. But you're, you're saying that the duration of the assets is, is longer than the duration of liabilities? Very long, because

- They've, they've invested a lot true to pension funds in equities Yep. And in foreign equities. Yep. And I'll talk about the, the foreign part as well. That's important because you get a lot of duration from the currency risk. I'll show you

- That. I think another thing that might be going on, I don't know how you, you measure or model the evolution of the liabilities, but I, my understanding is that their, their pension system has got a lot of macroeconomic indexation in it. It's got what in the, like German context we call sustainability factors, so that it does actually vary with macroeconomic conditions. But I So are, is the, do you think of the social security system just being like a fixed promise? Or is it, does it have all these bells and whistles?

- I think of it, I'm, I'm actually for the purpose of what I'm gonna talk about today, I'm not gonna get into that much detail in terms of exactly how the pension system works. I'll just show you the evolution of deficits and then show you how the picture changes once you consolidate all the different constituents of the public sector, not just looking at the central government. And then you realize that what they're really doing is, is taking on a lot of risk. And then the question is, well, why, why does that work? Can you really run a sovereign wealth fund with borrowed money? And I think that's where you need financial repression, because if everything is priced accurately in financial markets, then that shouldn't really work because your cost of funding is going to go up. So I'll talk about that, but I'm not gonna go into the nitty gritty details of exactly

- How, 'cause you're not going to just make one more comment, which is that it could then be yeah, that, you know, you're investing in equities isn't such a, as big a mismatch as you might think. It probably still is a mismatch relative to liabilities. The liabilities have macroeconomic indexation in them because those are also correlated then with market price. That's my, that was, oh, I see. That's where you come, yeah, I'll think about that. That's an interesting point. Not talking about that.

- So I'm actually curious about the political economy of how this came about in the sense of, like, I understand that you may not know, but do you have a sense of, you know, yes, I think I do. Who is affecting decision where and oh, to what extent? The, so Japan Central Bank is, okay.

- Yeah. So I, I'm, I'm gonna get into that. But the, the, the, the short story is there was always financial repression. The government had access to cheap funding. True. For example, the postal bank, which was the largest financial institution in the world, they were actually required to buy government bonds. And then in the eighties and nineties, they said, well, you have to liberalize. They liberalized remarkably, exactly when they fully liberalized, that's when the Bank of Japan started buying bonds. So they kind of replaced one cheap source of funding with another cheap source of funding. And I'll show you that the timing lines up nicely. So that is my understanding of what happened. Now, this is not a free lunch, because you have to think about how this impacts Japanese households, obviously, and I'll try to convince you that especially for younger Japanese households who are mostly saving in deposits, the Japanese investor is like a German investor. They don't invest in stocks, they invest in deposits. They're stuck trying to accumulate wealth at negative real rates. And of course, that doesn't work. They have very little duration in their portfolio. So when rates go down, there's no capital gains for young Japanese households. And so for them, it's, it's awfully hard to save for their retirement. So their large welfare losses for young non participants, which is a huge fraction of the Japanese population. And so the broader point here is it's not a free lunch. And in fact, if you think about it, it's a very regressive tax. It's a tax on less sophisticated, younger Japanese households.

- Bank of Japan increased the supply of reserves a lot. So I can, I know Japanese people are sort of forced to hold bank accounts, but how do they hold so much more bank accounts? Who's holding all the increase in reserves that has to be increased deposits.

- I'll show, I'll show you some numbers, but the, the household sector holds, I think roughly twice of GDP in deposits. So there's

- That, that increased during the period.

- Oh, yeah. Dramatically. I'll show you. Yeah, it went way up. Wow. Yes, yes.

- What what convinced them to triple the amount that sits in bank accounts.

- Oh, wait, that's a,

- With the facts. Okay, great.

- Yes, I'll, I'll get back to that. Okay. So let, let's just start by looking at this. The way we would typically do it, and this is sort of following the way, George Hall and Tom Sergeants think about the dynamics of the debt to output ratio. This is literally lifted out of their paper. The debt to output ratio today, you can write it as being determined by the past history of primary deficits, GE minus T, the cost of funding of the government R and the growth rate. Okay. And our T managed J to t, that's literally the cumulative cost of funding. So it's just the product of the one year cost of funding. And so that equation gives you a rough sense of how to think about that over output dynamics. And, and as you know, if you're in a steady state istic environment, so suppose that r and x, the growth rate and the cost of funding are constant, you're in a steady state environment where X is bigger than R, and I'm using X here, again, to denote the growth rate, to avoid confusing with G government spending. If you're in that region of the parameter space, and you of course abstract from risk, et cetera, et cetera, then you might say, well, the government can just roll over debts in perpetuity and run steady state deficits. But if you look at actual returns on outstanding Japanese government bonds, you see that that's not the case for Japan. The number we end up with is something like 150 basis points. The growth rate is negligible. Inflation is also very small, but, so we're definitely not in the region where the growth rate exceeds the government's cost of funding. So that's not it. We're not in that Goldilocks region. And so if you think about what's happened over the past 26 years, between 1997 and 2023, that's the sample we'll look at. Then you should have this number in mind. The average primary deficit was around 5% of GDP 26 years. That is roughly 133% of GDP in cumulative deficits over this period. So clearly not a sustainable trajectory there on, that's the picture you get when you just look at the central government. And that's what most people look at what we're gonna do. And that's really the main exercise here is just accounting, is we're gonna say, well, actually you should probably think about local governance, public pension funds, the BOGA, the Bank of Japan, the central bank, and publicly owned financial institutions because presumably there is a common budget constraint that all of these institutions jointly face. And then we're gonna analyze the budget constraint of the consolidated public sector. That's really the idea. And that's where the meat of the paper is. And, and this goes back to PA's question. Once, once you do this, the first thing you realize is that what happened is, prior to 2001, the Ja, the Japanese government had access to cheap funding because households were essentially trapped in low yielding deposits. There were interest rates, ceilings on deposits. Most households had deposits at the, the postal bank, as I said, that was the largest financial institution in the world, and they were required to hold government bonds post 2001. They said, hang on, we're gonna stop doing that, we're gonna liberalize. But immediately the Bank of Japan started large scale asset purchases. Of course, when they did that, they were mainly talking about the zero lower bound. But, but essentially, if you look at it from a fiscal perspective, it sure looks like what they were doing is substituting one source of cheap funding with another one.

- Yes. Is there a free movement of capital out of the country at this time? Or was it restricted in any way for households?

- Ah, I'll, I'll talk about that, but it, if you were a Japanese household and you would've wanted to buy US equities, that would've been costly and hard. Let me just put it that way.

- Okay. So, - But I'll, I'll come back to that in a second. Okay. Yes. Because you're, you're probably thinking, wait a minute, why don't they invest

- The No, especially young people today, they should just be buying the s and p 500 go to bed.

- Yes. No

- Double in five

- Years. That's certainly not what the Japanese, typical Japanese, I know, it's almost

- Incomprehensible.

- Yeah, yeah. In the

- Liberalization, what were they allowed to buy instantly? Postal savings banks.

- So if you wanted access to stocks, then you would've invested in these trusts that are essentially like mutual funds, but they're very poorly run and very expensive. And so you certainly would not earn the equity premium. You would earn the equity premium minus 200 basis points and

- The alternative source of cheap funding. So what went away? Yeah, so, so the households are saying, didn't have to just hold the deposits. I'm just curious what they were holding instead, then the bank, bank of Japan jumped into,

- No, they were holding deposits, but the postal office was not mandatory.

- Yeah, the, sorry, it should be cleared. The postal office was no longer required to, to invest in government bonds. I said, we're just gonna go to banks instead of the postal office. Okay, got it. Okay. So let's start before we actually do this consolidation by looking at the Bank of Japan's balance sheet. This is the Bank of Japan's balance sheet in 97, and everything is reported as a fraction of GDP to make it comparable across the different years. And, and as you can tell here, that's a, that's a pretty small balance sheet compared to the size of the economy. The picture below plots net purchases of Japanese government bonds for different sectors of the economy. I apologize, that's a bit small, but the key takeaway there is that the red lines are purchases by the Bank of Japan. And what you can sort of see is that around 2000, you see a massive increase in purchases by the Bank of Japan. The black line is issuance. And so that's not a mistake. The Bank of Japan was actually for a decade or so, purchasing more than issuance on a yearly basis. And what you end ended up with then is this balance sheet in 2021. That's of course a much larger balance sheet. If you look at the liabilities of the Bank of Japan, they now have roughly one GDP of reserves that they've issued. Right? So this is, these are reserves that you could think of just as deposits of financial institutions held at the Bank of Japan. And on the asset side, what you see here are the bonds that they've purchased by issuing these reserves that the bonds they've bought back. And so essentially, the Bank of Japan now holds roughly half of the outstanding stock of Japanese government bonds.

- What about foreign reserves? Were they, I don't see any foreign reserves anywhere near

- That's, that's in here as well in the background. I'll show you that in a second. This is just the Bank of Japan, but when I consolidate, that's all gonna be in there.

- Okay.

- So now, so, so here, this, this will have all of that. Oh,

- Yeah. Okay. I see.

- So now we're actually looking at the entire public sector. I first just wanted to show you what's happening at the Bank of Japan, because that's kind of a key driver. And so what you see here, if you look at the liability side of the entire Japanese public sector, is that before the Japanese public sector was funding itself, partly through these cheap deposits that I mentioned earlier, that was largely replaced by bank reserves issued by the Bank of Japan to the tune of one GDP. If you look at the asset side, that is also important. Of course, what you see here is that they have a domestic equity position of 31% of GDP, and then they invest half of GDP abroad without hedging the currency risk, importantly. So they're doing a currency carry trade. So these are really significant numbers, right? This is, think about it. So the equivalent for the US would be you're investing 15 trillion in, in foreign securities. They

- Said this includes the pension funds, right? It includes all that. What is the fiscal link between the government and the pension funds? How, how are they part of a consolidated balance sheet?

- Well, the, this is broadly speaking, a an unfunded pay as you go pension system. And so when there are shortfalls, so

- On top of the social security system, there's a pay as you go pension pension system. And they are the ones who are buying Ja, who are,

- Yes. But, but, but the, for part of the foreign reserves are also invested in, in foreign securities. Yes. But so

- Where, yeah, so the, are the unfunded pension liabilities a liability of the

- Government? Yeah.

- Okay. I mean, so those could be on the balance sheet, I suppose, right? I mean, I mean, if I were consulting the balance sheet, I'm putting the pension assets on there. Oh, I see. Yes. Actually, I should really have the total pension assets and the total pension liabilities on the balance sheet. If I'm, if I'm consolidating,

- Oh, I see you. Well, that's a bit harder because I then, then I would have to actually make projections about the size of these unfunded

- Liabilities.

- I didn't do - That. Well, you can make projections just of the size of the liabilities, you know, the present value liabilities. Yeah. The present value assets. The current value assets. That's, that's how I would consolidate a balance sheet with, with pension funds. It's like what we do with the US states, for example.

- Like That's a good point. We could do that. We, we haven't done that. But I think, so bear with me, because for the purpose of what I, the point I wanna make today, I'm not sure that, that, that would be, well, it might be relevant, but it's not material, I think to the point I wanna make.

- But these, these assets and liabilities don't add up, right? Because the, you got the, you got the bonds outstanding, but you don't have the present value of, of the, the payments and the taxes on the left side.

- Well, I don't have to explain that to you, but yes, if, if the Japanese public sector has a net liability that has to be backed by future surpluses Yeah. These don't add up to a hundred percent. So it's not a problem. Oh, no, no, no. The pension funds don't have the pension. Ah, yes. No, no, no. So you're right, they don't, yes, they don't match. Exactly, because they're, they're future surpluses. Let me get through that in a second. And pension liability. Yes. So yeah, what I wanna do first is go back to that first equation, but now I want to think about how the net debt evolves over time. So instead of just thinking about the debt of central, of the central government, I want to think about the value of debt minus the value of the assets, and think about how that involves over time. And then you see there's a second component there, excuse me, which is governed by the access return. The government earns over and above its funding costs RD on its risky investments. Okay? And if that access return is high, that's obviously, this is not, rocket science is gonna slow the growth of nd of the net liabilities of the government because you're harvesting these large access returns on your investments. And, and in fact, if that's big enough, you could end up in sort of a goldilocks situation where even though your cost of funding in a deterministic setting exceeds the growth rate, if that accessory turns big enough, you could, might still be able to run steady state deficits. So let me show you how big these numbers are. So the first thing we could do is go back to where we started. I told you that the total increase in outstanding debts as a fraction of GDP, if you accumulate all the deficits, was around 133% just for the central government. If you now subtract the excess returns that the government earns on its assets, you have to get to 70% because 70%, that's the increase in net liabilities, which is much smaller. Okay. So we started with 133% of GDP, it's down to 70%. That means that you can back out an excess return as a fraction of GDP of 2.4% of GDP. That's sort of the implied excess return. I'll, I'll do the actual return calculation on the next slide, but that's the number you should keep in mind. I'm still troubled by these pension funds.

- So pension funds are, they're, they're taking premiums and they're investing them abroad. And, and that's it. The government is issuing bonds, but the government doesn't directly own those securities. The pension funds aren't issuing the, issuing the law. They're not leveraging, I mean, maybe if they make all this money, the government can tax them at some point. But it's not clear to me why we put private pension funds in a consolidation.

- Oh, this is not private. This is is public. Sorry, this is all public pension funds. This is government run pension funds. There's no, oh, sorry, I should, yes. That this is all public, right? There's no private pension funds anywhere here.

- I see. And so to some extent they, they're, so they're, they're receiving premiums. Yeah. But they're also receiving government subsidies. Is that where it links with the taxes?

- Yes. And certainly there's a backstop right in the background. So,

- So we're just saying, look, you know, yeah, I'm, but then the premiums should be included as part of taxes, right? Or are they in your account? But these, these are largely pay as you go systems. So they're, they're, you're, so they're taking, they are measured in your taxes. 'cause those are payroll taxes that are, that, so you're saying those are, the premiums are included, which you're calling tax good.

- I'm not sure if it's entirely, it's, it's a, there's like three different tiers in the pension system. I'm, I'm not sure exactly how it's split out, but, but basically these are unfunded pay as you go systems. And we're just saying, look, there's a single budget constraint for the entire public sector. But yes, they're all public pensions. Sorry, I should have, yeah. Now I see where you were

- Confused when you guys do this with, with social security in the US context. How, how do you think of that? Do you, do you do like an accrued do, do you, do you, do you, in your other papers, do you, do you consolidate social security with the rest you do, right? The rest of the government budget. And so how do you, how do you do that? Because that's also an unfunded system. Yeah, and I, and I, and you know, my, my thought is, I mean, what you can do is you most, what what I understand some people is they say, okay, let's look at the accrued benefits, and then we have this, you know, trust fund, which kind of goes away when you consolidate it with the rest of the government government budget. But, but the, the accrued benefits then become a liability. Is that, do you have, are you doing that kind of, is it Well, yeah.

- Well we're, we're looking at the total deficits. So that's all in there, right? That's gonna be, it's gonna show up in the dynamics of deficits. But, but it's, but

- Yeah, he's not trying to take a present value that has a hundred and a hundred.

- Yeah.

- It's like there's this stuff that has to finance that flow.

- Yeah. I'm just doing simple accounting here and trying to get, you, give you a sense of how important these access returns are and that they harvest by taking on risk. And they're very important. So to their first order. So, because basically total debt increased by 133%. If you just accumulate all these deficits, 5% 26 years, but the net liability only went out by 70%. And so that means that they're harvesting around. If you believe this equation has a rough approximation, 2.4% of GDP. Now what I want to do is actually measure the excess returns directly. And then what you, what you get is a number that's pretty close, 2.2800000000000002%. So that back of the envelope calculation is actually pretty darn close. That's the point I wanted to make. How do we get to 2.2800000000000002% of GDP? Well, it turns out that the Japanese government earns an excess return of 190 basis points above its cost of funding on average. If you look at the last decade when they really got to work in terms of quantitative easing, that number goes way up, by the way. So if you think, as I do that quantitative easing lowers the government cost of funding substantially. This, you see that in that 4.66% number. The last column now expresses everything as a, as a fraction of GDP. So it computes the total dollar or yen amount, excuse me, in this case, and then divides it by Japanese GDP. And so you end up with a number, like 2.2800000000000002%, pretty darn close to 2.4%. Okay. So again, over the last 10 years, that number was much higher. It was actually 6.25%. So over the past decade from taking all this risk, the government harvested an additional 625 basis points as a fraction of GDP.

- Do they publish their portfolio?

- The the pension fund tells you roughly what they're, and we use that. Okay.

- What asset? But, but they don't tell you what assets, what, what stocks?

- There's, they're not Norwegians. They don't tell you exactly what they're holding, but they, but they do tell you broadly speaking, which asset classes they're invested in. Okay. Yes. So I think this number's pretty accurate. That's the Stanford Management Company. Yes. So these are break numbers. How did that happen? Well, let's go back to the consolidated balance sheet and think about how has that changed over time? So between 1997 and 2023, again, everything is expressed as a fraction of GDP. What the Japanese public sector has done is invest an additional 28% of GDP in equities, an additional one half of GDP in foreign securities. So again, these are large numbers. This goes back by the way to John's question. So obviously the sum of the assets is, let's say in 2023, 180%, but the sum of the liabilities is 275%. And so the difference 94%, you could think about that as the present of kind of value of future. Could

- They form this out to investment firms or do they do it directly the,

- Like - Goldman and JP Morgan and other stuff on the

- Foreign, the, the main Japanese pension funds? I'm not sure whether they use external managers. I, but I could look into that. That's a good question. Yeah, I don't know actually. I probably should know that, but I, I don't, but they have very specific targets for their allocations and they do publish those. So that's what, what's going on. Now, on the liability side, we've talked about this before. What they did was basically they borrowed at the floating rate over the last decade by borrowing one GDP at, at the floating rate, which is the rate they pay on reserves, which actually throughout the period was essentially zero.

- So you're saying that their debt to GDP ratio is really only about 68%, not 250.

- That was your bottom, the bottom For the consolidated public sector, it's much smaller. Yeah. So that's correct. The, the

- Issue I have with that is I think that when you include the fact that they've borrowed a bunch of money from employees through pension arrangements, that's a huge liability and that's gonna increase the liability back. That, that's what I'm trying to get at with asking about the unfunded liability of the Yeah, total liabilities, not unfunded, just the unfunded total liabilities on the liability side of the balance sheet. Because that's a, that's a source of bar, that's the way you think about the

- Pension. Yeah, I'm, I'm trying Yes, I agree. I mean, if you wanted to quantify that, then I should definitely do that. The question I first wanted to address as a simpler one, people look at this and they think, how is this, how were they even able to do this without getting massive inflation, for example? And I'm saying, well, actually if you add it all up, their are smaller than they appear because they harvest these large excess returns from their risky investors. But I know, but again, that I'm not saying that that's a good idea, right? Yeah, yeah. That's where I'm going. My

- My, my question is then, so the returns

- Yeah.

- From these invested assets Yeah. Doesn't go to the pension year. It's just the way, like, you know, I'm trying to figure out, you could imagine yeah, if I'm running this

- Yeah.

- As a private pension fund, you know, you have the allocation and then the allocation will go to this is a pay as you go system.

- Yes.

- And it's, it's a defined benefit, right? So in some way they're, they're doing like an insurance company. And actually in Italy they did. Exactly. Yeah. But there was financial repression, but it was done to insurance companies in which the insurance company was, were mandated to hold it by the government to hold government bonds.

- Yes. - So in some way that, that was the, so in, and, and then the question is, you know, the, the returns who do, so basically there is a redistribution of, of returns. There's a lot of re in the, in the back of these, yes. In the, yes. Absolutely.

- Another way to put it is the pension system, rather than ours, which buys government bonds. There's, but foreign stocks, they made a killing on the foreign stocks. Those can finance a large proportion of the pension payments. And so only 60% of GDP is left over to be finance.

- Yes. But this, this, this financial engineering wouldn't work on without financial repression. That's what I, I don't, I don't agree with that. I'm

- Gonna get, because the unfunded liability, he added the assets to the balance sheet, but not the liabilities to the balance

- Sheet. Those are constant, those didn't rise. The

- Liabilities are constant. I don't think that's

- True. We're going back to the last slide here. When the stock market went up Yep. They, the pension funds made a killing. Their liabilities didn't change.

- Oh, their liabilities absolutely did go up by a lot because interest rates came down by a lot. So the market value, their liabilities went up by a lot. Maybe they didn't market on the balance sheet, but they absolutely went up tremendously in value.

- Yeah, we, let me, let me, I'm gonna think about that because I think that's an important point. So be, but just to summarize what they've done, what they've done is they've taken a lot of risk in equity markets and in currency markets. So they've invested half A GDP in foreign securities. That by the way, makes them essentially the largest currency carry trader in the world because what they're doing is they're borrowing in the end a low rate funding currency. In fact, when you explain the currency carry trade, that's typically the example you give. And they turn around and invest abroad in higher interest rate currencies. Okay. Now let me, let me do a few counterfactuals and this will set us up for a discussion of financial repression. Suppose that there had been no qe Japanese sources estimate that the effect of QE was really large, that they, that that lowered yields by roughly 200 basis points cumulatively after 2012. Suppose we actually run that counterfactual. So suppose that the returns on Japanese government bonds would've been 2% higher than the actual returns. What would be the effect on these access returns? Well, it's pretty dramatic as you can see from the bottom panel. So that's the no QE counterfactual. And in the last column you see that now the, the, the sort of the average return as a, as a fraction of GDP goes down from 228 basis points to 66 basis points. What if they were actually forced to hedge the currency risk? Which by the way, if you're a Japanese financial institution investing abroad, you're, you're sort of forced to at least hedge part of the currency risk. You can't just leave all that dollar exposure unhedged, if you were to force the Japanese public sector to hedge the currency risk, that also dramatically lowers these returns. And if you combine all of it, then the entire excess return is gone. So to understand why they're able to do this, you have to think about QE and the fact that they're taking on a lot of currency risk and doing a carry tree. If you take that off the table, then the magic is gone. Okay. Excuse me. There's a question. I think.

- Yes, please, Bob McCauley here. I wonder whether the long foreign currency position can be regarded as a hedge rather than an outright risk. The Japanese government bears both the earthquake risk and the international security risk. And either one of those could involve substantial imports upon realization, indeed, the latter risk security risk, they're already stepping up their defense spending, much of which will go for foreign weaponry.

- Thank you. That's, that's an interesting point. Honestly, I hadn't considered, you're saying this could be the Japanese public sector hedging some dollar exposure that it already has. I'll, I'll think some more about that. But for the purposes of today, all I want to do is point out that this foreign currency exposure is a big source of the excess returns that and quantitative easing combined account for pretty much the entire excess return.

- Paul, following up on that thought, yes. The, the is at a very weak level historically in real terms. So it's interesting to consider what mean reversion might imply.

- Yes. So I'll, I'll get to that. This is another footprint of financial repression over the past. So 25 years between 97 and 2022, the real exchange rate has depreciated by roughly 50%. Why does that happen? Well, if you artificially lower the real rate of return on your investments at home, that that is sort of what almost any model of exchange rates will tell you has to happen. Your currency has to depreciate. And that's happened in a big way. That's another cost of financial repression. If you think that's what's going on, what is financial repression is just different ways the government uses to lower its cost of funding. This is something that even the US has done on occasion. If you look at the Civil War, the First World War, the Second World War, and arguably maybe even COVID, although that might be more controversial, there's some work by George Hall and Tom Sargent that argues that that is the right way of thinking about what happened there. Even though they don't necessarily use the word financial repression, I think they certainly suggest that what happened during these big wars and what happened during COVID was, was similar and, and part of the government's attempt to lower its cost of funding. I've already sort of given you a little bit of the institutional detail. Let me actually then get to what I think is a key question. Is the debt priced correctly? So why is that an important question? Well, ultimately you'd wanna know, can you really do that, run a sovereign wealth fund with borrowed money while keeping the debt risk free? And I think the answer there is, well no, on the asset side here you have, and this is what John was sort of complaining about that I missed on the earlier balance sheet. You want to have a claim to government surpluses, right? And then you have of course your financial assets, all the equities you bought, loans you've invested in foreign assets. That's your asset side. What do you have on the liability side? Well, that's what I just showed you, mostly reserves and government bots. Okay. Alright. Now basic finance, no arbitrage, this sort of probably reminds you of Milani for a company, right? If you look at a corporation's balance sheet sort of says, well the expected excess return on the right hand side has to equal the weighted excess return of the surplus claim and the risky investments, that's just standard. No arbitrage. The ratio of risky assets to debts is roughly 66%. And so what you then can wonder is, well if we, if we sort of do the math, do we, what do we get? And I think what you learn really quickly is that debt is probably overpriced if you do the numbers. And here's why. So plug in that 0.66 number for A over D. What do we know? Well we know that the realized return access return on risky assets was roughly 190 basis points. And so if you rearrange that equation, you end up with an equation for the spread between the return on your risky investments RA and the return on that surplus claim. What is that surplus claim? Well that, well that's what's in the top left hand corner over there, right? That's a key part of the balance sheet. Yes.

- I gotta object 'cause there isn't arbitrage. And another way of saying it is if the government has, if you on the left side of Laffer curve, it's certainly possible that fiscal surpluses will always make up the difference. In fact, it's al it's almost an accounting identity that they, so Oh yeah, yeah. It'll make up the difference if you include the inflation tax.

- Sorry, that's not where I'm going. I'm, I'm just talking about, so I'm making a slightly different point. I agree with you. That's not what I'm gonna say. I'm gonna, I'm gonna argue that what this implied is that the excess return on the surplus claim has to be deeply negative, has to be ver So it has to be a very negative beta security. If I were to secure, alright? Yes. But that's hardly in line with the data. Well, it is possible surpluses are not strongly,

- Well if they do not default it or inflate it away, then they will have to raise taxes or cut spending.

- Yeah.

- And then you know it's gonna happen.

- Yes.

- You're just saying they're likely to inflate it away.

- If you look at a picture of surpluses. So here's what John and I are, are talking about. So not to date, but they, you know when, when the crisis comes or after taxes? Yes, yes. It's, it's actually very hard to find examples of fiscal crisis where all of a sudden you, you sort of see a radical switch in the, the cyclical properties of these sur surpluses crisis isn't cyclical,

- A crisis goes boom.

- Yeah. This is about the, this is about the risk properties of, so suppose we were to trade a claim to surpluses. What this math says is it has to be a very safe claim, which means that you would run huge surpluses in recessions. That's not what the Japanese government does.

- You're mistaking the current surplus and the present value surpluses. No, no, no, no recessions. We always lower current surpluses,

- But raise the expected future one. What you're saying is, well I could, yeah, this stock looks risky, but actually, you know, look, going forward, actually it could be completely countercyclical negative beta. That that, that doesn't seem like a compelling argument. Yes, it could happen, but it doesn't seem very likely. And there's, I cannot think of a single example of an advanced economy where governments do that. Governments don't run large surpluses in recessions. And this is what that says. That's all you were speaking the surplus in the present value of the surplus. No, no,

- No, no. John I think goes down recess. But the present value on unaffected, because they borrow now and they pay it back later. This is, this is about the risk properties of the surpluses. Right. Which is about the total rate of return. I understand. 'cause the current payment and the present value of the future. So the current deficit, yes, can co, can, but that's, we're very happy with the rise in the present value future ones. But that's,

- We're gonna pay it off. So we, bonds always work. Ultimately that's all governed by the, the properties of the cash flows. Right. Which you are assuming are an

- A one and they're

- Not. No, I'm not, I'm not making it. I'm just, look, this is what it is. That's just the math You could say John is saying, I think that's entirely reasonable. I think it possible. Okay. Yeah, I no, I, I certainly wouldn't want to argue that. If you were to say, okay, if you were to come to me, you're at Goldman and you told me Hanno, do you wanna buy an ETF? That's a claim to Japanese government surpluses. And you're gonna tell me the expected excess return is minus 2.96%. I think I would pass. That's, that's all. But anyway, let me, let me skip this. This is, I think, so if you do the math, you get to a wedge of 180 basis points. So this is saying the debt is, the yield on the debt is too low by roughly 182 basis points, which is in line with the estimates that I've read of the effect of quantitative easing. Where does that show up? So some sort of utter evidence is, and I think this was mentioned earlier in a question in the currency market. So over the past 27 years, the real exchange rate has depreciated by 45% or 1.6800000000000002% per annum. Okay. And so another way to get at this, sorry Ryan,

- You had a question? Oh yeah, I'll wait until

- You're, another way to get at this is to say, well, to make sense of what's happening to the exchange rate, we could think about something like long run UIP long run. UIP says, well if you take a, a 30 year US bond and a 30 year Japanese bond, let's make it real. And you compare the yields, then that tells you what the expected rate of depreciation of the yen in real terms is against the dollar. So looking at the long end of the yield curve. Okay.

- Can I ask just Yes, like backing up a little? Yeah. Does this, you know, you're saying this is a, which I think is right in those obviously have the, you know, they take the stairs up and elevators down.

- Yeah.

- Just like what is your view of this strategy broadly? 'cause two to 3% of GDP is definitely very large amount. Obviously there's costs and you're gonna talk about Yeah. You know, to younger households, but it, it doesn't, you know, they're able to do this consistently over the course of two decades. Well, well, very

- Well and earn a lot of is I think the wheels are coming off this experiment in the sense that they're well away. They're pretty far away from the zero lower bound. Now inflation is running at 3%. And so it looks like what's going to have to happen is that the entire yield curve will be pushed up. Now given the duration mismatch that's built into this exercise, that means that that would destroy fiscal capacity dramatically for the public sector. So that would be what, what ends this experiment? Yeah. So that's the sense in which they could get into trouble. But as, as Paola said, this also has very unappealing cross-sectional implications. So let me, let me actually, I think I'm almost out of time. So let me briefly go there. But basically from long run UIP you would infer a wedge of roughly the same ballpark. Okay, so here's the exercise. You would say, well wait a minute, in order for this 45% real depreciation not to happen, how much higher would long-term real yields have to be in Japan? And you get a number around 180 basis points. So that looks like a plausible number. That's the point that I want to make some independent evidence. So, so your colleagues at

- GSB might say that this isn't financial repression, it's just the wonderful liquidity value of government bonds.

- Yes. I didn't mean to sound that one person's convenience fields to another person looks look like, looks like financial repression. I don't think that those Japanese depositors are super enthused with all the liquidity and safety they're getting. You can tell, but it could be true. And it's hard to refute that. Right.

- Also now that huge depreciation of the yen, that's, that's, that's a lot of inflation built up when they catch up to international prices, right?

- Yes. And, and I, that's actually, I think some of the most compelling evidence that this is really, it's all about distorting long-term real rates. This, this sort of huge, massive real depreciation I think is because

- Stuff is really cheap in Japan and when they start paying our

- Prices. Yeah. So

- They, in that, that's a builtin inflation as we finally get back to ppp, right? Yes.

- What's the longest dated dollar you end futures market? How far out can you buy?

- How does it

- A year took like a year. I don't see anything longer.

- Okay. So don't live that long out there. So you

- Don't live that long.

- I I cer certainly five years. I, in fact, I'll show you some. So real, real quickly, and I I think this also came up, maybe it was your question or your question I've forgotten. Why don't Japanese banks or intermediaries, investor abroad, it's hard for individuals to do this. Well, they have to hedge to currency risk and maybe not all of it, but certainly a significant chunk of it. Otherwise regulators would be on their case if they end up with trillions of dollars in dollar exposure. What's the result of that? Well, because of all the hedging pressure, because there are all these Japanese investors who are trying to do just that covered interest rate parity breaks down and you have very negative basis, which acts, the way you can think about it is it's a tax on foreign investment if you're hedging. And these taxes are significant actually. So on average they're between 50 and a hundred basis points. So in terms of maturities, this is one month, three months, six months, one year. Sorry. The, the axis here are very small. I apologize. This is for longer maturities. So here we're going out to Chris, can you tell what the longest maturity is on there on the bottom? I can't see from here. 30 years. 30 years. So you can go out pretty far. Yes. So huge negative basis everywhere. What what does that mean? Well, there are all these intermediaries, they're intermediating 3G DP and deposits, insurance, pensions, they want to invest abroad, but they can't all do it at the same time if they wanna hedge, because then that creates this negative basis, which is like a tax. If you're the government, you look at this and you might think, well there are ways I can fix this by changing financial regulation. But it's not clear you want to, because this helps you trap

- Japanese,

- They're all criminals.

- The government is, is a criminal organization basically.

- Well, I I'm not gonna go that far, but, but, but certainly this is not in the interests of, of the,

- Of the, of the

- Public. No, it's not. I think that's correct. That makes it, yes.

- Alright.

- So, so if, if you're thinking, how do they get away with this? That's part of the answer is that they, they have these CIP deviations that actually work in their favor and, and actually make this somewhat more sustainable. Okay.

- You're also earning the equity premium.

- Yes. But, but still, you know, you're, you're giving up Yeah. Between 50 or a hundred base points. And if, if these flows would be much bigger, presumably that basis would go up. Right. It's, it's, it's,

- But why don't our, I mean our hedge funds have been doing this for a long time. Bowing yen investing in dollars and

- Well, I think, right, so the way to think this is the Japanese government does the carry trade, but they don't let the private sector do it.

- In Japan.

- In Japan, correct. But the private Goldman Sachs can do it. Yes. But are, are you, are you referring to the carry trade now or to the CIP deviation? Sorry, I I, or both? Both. Well

- Carry trade need hedging.

- Yeah. Okay, go ahead. Yeah, the carry trade is so, but the key, the key point is the government doesn't hedge the private intermediaries. Hedge. And that's, and then they pay a tax because that's,

- You were talking about repression.

- So - Risk regulation says you guys gotta hedge.

- Yes, yes. That's the repression part, I think. Alright, so I'm, I'm almost done here. The, the last point I wanna make is, well actually if you look at their net liabilities, and this is a little counterintuitive. The duration of the risky assets is quite long because they have all this equity, not trivial to measure duration of equity, but, but it's a large number. It's roughly the price dividend ratio, that's roughly how you measure it. And so if you add it all up to duration of the risky assets is around 29, 30 years, but duration of liability is much smaller. And so even though they have way more liabilities than assets, the asset duration component dominates basically, and so they end up with a negative duration number. What does that mean? Well, it means, it means, well if the duration of your liabilities is smaller, then the duration of your surpluses and the surpluses are way out in the distant future. Then if you lower real rates, then yes, that creates extra fiscal capacity. Of course, conversely, if you increase real rates, then you're destroying fiscal capacity. Why do lower real rates increase fiscal capacity? Well, you know, the government's gonna be borrowing for a long time until they start generating surpluses, and so they're now borrowing at a lower rate for a longer time. So that's what's going on on the government side. By the way, if the government wanted to be perfectly hedged against interest rate, risk hedging came up earlier for the government, then what you'd have to do is match the duration of your liabilities to duration of surpluses. Right. So that would mean you would borrow at very long terms that that's not what they do.

- I would think there'd be huge pressure in the Bank of Japan to hold interest rates down, especially long term rates. Just, this is sort of the new symphony of, of the world is

- Yes,

- We want low interest costs on the debt and we'll see when the inflation

- Crops. Yes. And, and central banks have invoked all sorts of plumbing problems that have appeared recently all around the world to explain why they cannot possibly shrink their balance sheet dysfunction. Yes, all yes. So I, I'm sympathetic to that few, so I'm not gonna argue with you. Well, the

- Interesting open question is how long does it take to get the inflation, which we don't really know.

- Well, we'll find out. Yeah. But Japan now has 3% inflation. Yeah. It's I think roughly at par with the US and they're still buying bonds. So, so we talked about Japanese households being trapped in deposits. John, I think this was your question. Yeah, yeah. So now here you see a juxtaposition of Japan and the us so it's quite dramatic, right? So Japan, the ratio of deposits and currency to GDP is now 190%. In 2021, it was actually two, that's the number I gave you earlier. And so much higher than the US and, and obviously a dramatic increase between 97 and 2023. Right. So households do seem trapped in deposits, and I don't think it's, it's because they're happily enjoying the safety and liquidity of deposits necessarily. It might just be a lack of viable alternatives, a lack of financial sophistication, all these sorts of things that we usually don don't think about.

- But on it, would it make sense to think, I mean, we had this conversation with John the other day, but would it make sense that to some extent there is a, you know, of course there are distribution effect of this, but the, the, the Japanese households are partially funding this operation unwillingly, maybe

- Yes.

- Versus the American one. They're not. So in some way it's more fragile when you actually are opened this, it's almost like a, a closed economy in terms of investing, investing abroad. Right. And so in some way they can carry this game for a longer time. I'm not justifying it, but to some extent it, it makes it less

- Volatile. I agree. In fact, I, I think it'd be very hard for the US to engineer a similar type of financial repression. And, and I mean, the first line already shows you it's gonna be hard, right. Americans are not as enamored with deposits as Japanese or German investors.

- These equities have to include 4 0 1 Ks. Right. So we have, we have equity investments in our pensions

- Yeah.

- Which we're allowed to do. And it sounds like they don't, they have defined if you don't have defined benefit pensions.

- Yeah, yeah. And, and that, and so, and actually if you look at participation rates, what you can see is yes, Japanese households have deposits there. I think maybe they, the participation rate is even higher than for Americans, I think.

- But once you also saving rates are much high,

- Are very high, much, much higher. Yes. Yes. But if you look at securities then, then you realize that participation rates certainly in the lowest income deciles are quintiles, excuse me, are are very low, right? Yeah. They're, they're super low. Basically most Japanese households are, are not participating in, in any asset markets. Yeah.

- If you, if you update the fed the flow of funds stuff to 2025 with the latest data that comes up, pretty often it's gonna, it's gonna look the difference between the US and Japan is gonna look even bigger. Yeah. I mean just there, there was that recent post I think you saw where, you know how household us household wealth is now eight, eight times average income or, and a lot of that has to do, I mean this thing correlates with this like high beta, that numbers.

- So if you think about, this is the last thing I, I wanna mention. If you think about the sort of cross-sectional implications, then let's take a, just to fix ideas, let's take a young Japanese household that is mainly invested in deposits. So they have no duration on the asset side, but they have to save to finance their consumption after retirement to, to the extent they can't fully depend on the government program. Well, when rates go down, they don't get a capital gain because they have no duration and, and, and so now they're saving at much lower rates going forward. So they're worse off. And in that sense, financial regression you can think of as a regressive tax, you can actually quantify this. And the way you quantify it is, and this is similar to what we did for the government balance sheet, you think of a household's financial wealth, let's leave housing aside for a second. Just as a present discount value of its consumption minus its income, you can come up with a duration number for financial wealth. Obviously if you have only deposits, then that duration number is close to zero on the right hand side. That is determined by the profile of consumption relative to income. But the point is, for a young household, the duration of consumption minus income is gonna be pretty high. And then when rates decline, if you don't have a lot of duration in your financial portfolio, but your excess consumption, excess consumption relative to income is pretty far out in the future, then when rates go down, you're in case B and your consumption possibility sets shrinks. You're not gonna be able, in other words, to afford the same amount of consumption, you're gonna have to cut back when real rates go down. On the other hand, if you're a financially sophisticated household and you have a lot of stocks and bonds, let me actually show that here. That's case A, then you have a lot of duration in your portfolio and then you're more than hedged. And then when rates go down, your capital gains are more than enough for you to be able to finance the same consumption stream as before.

- Are capital gains taxes lower than ordinary income?

- That is a good question. I'm gonna have to look into that. I don't know. That's a good question. That is relevant here. Yes. So basically if you're a young Japanese household and rates go down, you're, you're sort of worse off, especially if you're saving in deposits. If you're an older, more sophisticated household, you're better off because the capital gains are more than enough for you to keep financing the same consumption stream. And so if you compute welfare costs of 1% decline in interest rates, you, you actually get pretty big welfare costs for non participants. And again, this is all governed by the difference between the duration of this consumption you need to finance and the duration of your portfolio. If that gap is zero, then there are no welfare implications. That's basically the bottom line is that, and this is what I was saying before, that financial repression is sort of like a regressive tax because as your lowering rates below what they would be, younger households are gonna have to cut back on consumption, especially if they're saving in deposits and they're less financially sophisticated. I'm almost out of time, so let me actually conclude here by sort of giving you maybe a couple of big picture remarks. Traditional macro view, and I'm painting with a broad brush here when it comes to quantitative easing and real rates, is that advanced economies are experiencing demographic transition, secular stagnation, all these forces lead to lower equilibrium, long run real rates that creates extra fiscal capacity potentially. And as economies bump into the zero lower bounds, central banks deploy large scale asset purchases. I want to present an alternative view, which is yes, advanced economies are experiencing demographic transition, secular stagnation, but these forces directly lead to large government deficits. And that is why governments resort to low rate policies in order to be able to keep the fiscal ship afloat. And central bank might help, central banks might help by deploying large scale asset purchases. These two views are potentially complimentary, right? It's not like there is one right view, but I think this one certainly has some merits, especially when you look at economies like Japan. This view has very different implications when you think about welfare and, and the heterogeneity of financial sophistication. Duration in household portfolios, I think is a, is an important topic that we need a lot more work on. When you lower real rates, you're raising the prices of all financial assets and other assets with a lot of duration like housing. And that has important cross-sectional implications that are typically ignored when we think about these policies. Let me conclude. Thanks for all your comments by the way. Let me conclude. What we did basically is just an accounting exercise. We consolidated the Japanese government's balance sheet. We tried to argue that the Japanese government has been engaged in risky maturity transformation, co financial repression, and that they're running a very large currency carry trade. Again, with financial repression, keeping the government's cost of funding low. I think it's artificially low, but we can, we can debate that. Is that a good thing? Well, it depends on which way rates go. If rates go up, the maturity mismatch on the government's balance sheet would mean that fiscal capacity gets eroded very quickly. This also has important implications for welfare because it all depends on whether you have duration in your portfolio. I think. So that's, that's broadly what I have.

- Thank you. Not a reliable picture of a defense partner. Not a reliable, where are they gonna get the extra money to raise share? I already made that point too. We're not a

- Lot better shape. Alright. Thanks so much, Hanno. Thank you.