What would Japan have to do to achieve a 10% inflation target?

I do not expect Japan to adopt a 10% inflation target, nor do I think it’s a good idea. Nonetheless, it’s a useful thought experiment, which helps to illuminate some difficult to grasp ideas.

Let’s think about what Japanese interest rates and base money demand would look like with a 10% inflation target, assuming no IOR. Here are some plausible guesses:

1. Japanese interest rates would be roughly 10%, plus or minus 2%.

2. The Japanese monetary base would be roughly 5% to 10% of GDP.

I base the first estimate on the fact that real interest rates in Japan are currently close to zero, or perhaps slightly negative. With 10% inflation, real rates might fall slightly (due to the zero bound being lifted), or rise slightly (due to the taxation of nominal interest.) But history suggests that higher inflation expectations are mostly passed along as higher nominal interest rates.

One the second point, 5% is roughly the figure I’d expect, but given the unusually large Japanese demand for currency I put in 10% as an upper bound.

Now let’s think about where Japan is today:

1. Interest rates are roughly 0%.
2. The monetary base is roughly 100% of GDP.

Thus, to achieve a 10% inflation target the BOJ would have to raise interest rates to 10% and then cut the monetary base by 90% or 95%. After that initial cut, the base would rise by roughly 11% per year, assuming 1% trend RGDP growth and 10% inflation.

Now here is where things get confusing. Just raising interest rates to 10% and cutting the base by 90% will not magically produce 10% inflation; just as having millions of New Yorkers leave their house with umbrellas will not magically produce rain. It’s more the other way around. If you seed the clouds over New York to produce rain, then millions of New Yorkers will leave their house with umbrellas.  NeoFisherian meteorologists claim that you create rain by ordering New Yorkers to carry umbrellas, citing “equilibrium conditions”.  Market monetarist meteorologists claim you create rain by creating conditions where millions of New Yorkers choose to carry umbrellas.

Do you see the subtle distinction? It’s true that a non-coercive policy that indirectly causes millions of New Yorkers to choose to carry umbrellas will also cause rain, but a coercive policy of umbrella carrying will not work. Thus seeding clouds will cause rain, but having a crazy dictator say he will execute New Yorkers for not carrying umbrellas will not cause rain.

With monetary policy, you cause 10% inflation expectations with a credible policy to buy as many assets as needed to move market inflation expectations up to 10%. When is a policy credible? When the government actually intends to carry it out, whatever it takes.  In that case, you probably don’t have to buy anything. Indeed just the opposite—you reduce the monetary base.

A few years ago, the Japanese government adopted a 2% inflation target, but markets correctly understood that the Japanese government had no intention of doing whatever it takes to carry out the policy. And they were right. If the Japanese government had intended to do whatever it takes, then markets would almost certainly have seen that. Markets are not stupid; they can pretty easily see whether governments are serious or not. Or if they can’t, it doesn’t take long to figure it out.

Governments are like 3-year old toddlers.  They don’t have a poker face.  Their emotions are clearly written in their facial expression.

To summarize, asking whether QE “works” is like asking whether umbrellas imply rain.  It’s a meaningless question, unless put into some sort of context.

PS.  Try this:

1. Do umbrellas keep people dry?

2. On average, are umbrella carriers wetter than those not carrying umbrellas?

I don’t know about you, but on average my pants get wetter on days that I carry an umbrella (due to wind) than on days I don’t carry an umbrella (and it’s usually not raining).  So do umbrellas “work”?

QE is aimed at leading to faster NGDP growth.  But, on average, countries doing QE have slower NGDP growth than countries not doing QE.

Hospitals are aimed at keeping people healthy.  On average, however, people in hospitals are less healthy than people outside of hospitals.

Unfortunately, all these metaphors fall short by ignoring the role of expectations.  But it’s the best I can do.



23 Responses to “What would Japan have to do to achieve a 10% inflation target?”

  1. Gravatar of Paul Paul
    2. June 2019 at 19:21

    “Expectations” is a weak channel for monetary policy.

    I always read blogs hoping to gain some more knowledge of why the money supply and nominal GDP are the only thing we need to understand business and financial cycles, but I’m always left seeing monetarist economists as the meteorologists who demand we carry umbrellas or they’re going to make it rain.

  2. Gravatar of ssumner ssumner
    2. June 2019 at 21:32

    Paul, I’d read the post again, I think you missed the point.

  3. Gravatar of dtoh dtoh
    3. June 2019 at 03:46


    I think everyone should take a step back and look at what Japan has actually done.

    They’ve exchanged assets (reserves) for other assets (mostly JGBs) with the banking sector.

    Transactions that take place wholly within the banking sector have no impact on monetary policy. Increases in the base have no impact if it’s a merely an increase in reserves held with the CB.

    For monetary policy to be effective the CB has to exchange assets for money with the non-banking sector. They have to get the non-banking sector to hold more money. If they did that, they could generate 10% inflation tomorrow (or at least by the end of the week.)

  4. Gravatar of bill bill
    3. June 2019 at 06:33

    Wonderful post.
    When you wrote “it’s a meaningless question” at the end, it reminded me of Marisa Tomei in My Cousin Vinny. “It’s a trick question”.

  5. Gravatar of Matthias Görgens Matthias Görgens
    3. June 2019 at 07:49

    Paul, expectations are an incredibly powerful channel.

    But expectations require credibility. And they can only bring the effects of the future closer in time, anticipating them. You still need an actual channel (even if it’ll look like that channel wouldn’t be doing anything, if it was completely anticipated).

    Scott, apropos expectations. It would be interesting to speculate how this would play out if the Japanese central bank actually had unlimited credibility in creating inflation.

    The day they announce (or just leak!) the policy inflation would pick up, and people would start returning cash to the bank?

    Bank reserves would pile up. But the government bonds on the central bank balance sheet would also lose in value. Even in nominal value, because interest rates would pick up. Or what would happen?

  6. Gravatar of Michael Sandifer Michael Sandifer
    3. June 2019 at 08:15


    Expectations are the only channel for any policy. If people don’t believe a policy will work, it won’t. There won’t even be an attempt to make it work, by definition. Markets react strongly to Fed comments for a reason.

    Expectations determine velocity when it comes to monetary policy. You can define velocity as 1/d, were “d” is the demand for money.

    If you want to effectively criticize market monetarism, focus on the reliance on unobservable variables. Expectations are obviously central in all social sciences, so attacking the expectations channel will get you nowhere.

  7. Gravatar of Michael Sandifer Michael Sandifer
    3. June 2019 at 08:24


    I’m curious as to why you want to still use metaphors that only partially fit so many years into your public education effort on monetary policy. Why not just discuss the explicit variables, as you have on occasion? Is it because you’ve found the latter approach isn’t as edifying? This is not a criticism, but a question.

    For example, why not talk about the demand for money affecting velocity, which changes the neutral interest rate, etc.? It just seems much clearer to me, but that might be my bias. You have many years of experience teaching economics to both college students and the general public.

    I formulate it like this: change in NGDP = [(1 + change in Sm)/(1 + change in Dm)] RGDP,

    where Sm = supply of money and Dm = demand for money.

    For nominal shocks, the short-run change is to nominal GDP, the bit longer term change is to RGDP, and then wages adjust and money is neutral in the long run.

  8. Gravatar of Benjamin Cole Benjamin Cole
    3. June 2019 at 09:08

    I am beginning to wonder about this public expectations jazz in relation to inflation.

    What public? Kim Kardashian has 60 million followers on Twitter. That’s nothing; a singer named Rihanna has 90 million.

    For all the flaws of the financial media in the US, it is perhaps the only forum where there is even a discussion of monetary policy. I would say the financial media for the last 40 years has dourly warned about the impending outbreak of hyperinflation. For that matter, at the Cato Institute they are still predicting much higher inflation and interest rates just ahead.

    Almost nobody predicted or expected lower inflation and interest rates as a secular trend. That is, nearly the entire orthodox macroeconomics profession has not made the correct prediction of a long-term secular trend down in inflation and interest rates.

    Such luminaries as Paul Volcker and Martin Feldstein have repeatedly warned that much higher inflation rates are right around the corner.

    Perhaps Kim Kardashian and Rihanna should be asked by the Bank of Japan to inform their followers that an outbreak of higher inflation is right on the horizon.

    David Glasner says the explanation of public expectations in determining inflation is something of a tautology.

  9. Gravatar of Paul Paul
    3. June 2019 at 12:39

    I think I know plenty well what you’re saying, Scott – it’s just wrong. In a world where money is entirely outside the CBs control (except the quantity of base money, which doesn’t – contrary to textbook mechanisms – “multiply” into broad money), a rules-based, monetary policy-only regime is not feasible.

  10. Gravatar of Christian List Christian List
    3. June 2019 at 15:24


    Your metaphors are legendary. They don’t always help to make things more clear but I love them nevertheless.

    Now kind of OT (or actually not).

    Few days ago Kuroda said:

    “Our estimates roughly show that the chance of inflation reaching 2 percent during fiscal 2021 is low.” (Very convincing, Kuroda, omg!)

    And then: “It’s regrettable that inflation has not reached 2 percent for six years.”

    –> Is this guy a comedian, or what?


    And just 5 days ago Reuters reported:

    “Bank of Japan Governor Haruhiko Kuroda said major central banks may have to become more flexible targeting inflation, as they are missing targets due to the price dampening effects of technological innovations and globalisation.”


    Becoming a Fed nominee is not enough, Scott. They desperately need you at the ECB and at the BOJ, too.

  11. Gravatar of dtoh dtoh
    3. June 2019 at 16:29

    One more thing. Japan’s inflation target is completely misguided.

    1. Monetary policy is only effective at raising output when a) the economy is not at full output, and b) you have sticky wages and prices.

    2. Given Japan’s tax and regulatory regime, the economy is at or very close to full output.

    3. After 30 years of a stagnant economy, Japan’s wages and prices are not very stickuy. Most employers could easily cut wages if they wanted.

    4. The reason Japan’s economy is not growing is because of confiscatory tax rates and suffocating regulation. Talking about an inflation target or monetary policy distracts the public and the policy makers from the real problem.

    5. If Japan fixed its systemic problems (tax and regulation), the economy could easily grow at 5% or better for at least the next ten years.

    6. It will never happen because there is zero recognition of the problem within Japan.

    7. Japan is doomed to a slow decline for at least the next two to three decades.

    8. In the meantime, it’s a very pleasant place to live or visit (as I’m sure Rome was before the Visigoths showed up.)

  12. Gravatar of ssumner ssumner
    3. June 2019 at 18:58

    Michael, I’ll do that in the book I’m working on.

    Paul, If you actually understood what I said and still wrote that comment, it’s much worse. Your comment doesn’t fit my post at all, even as criticism.

    And there’s nothing wrong with the textbook model of the money multiplier, if you are referring to Mishkin’s textbook, which is the best seller in the field.

    If you were correct, then the Japanese could buy up all of planet Earth with zero inflation. They’d live off the hard work of foreigners. Wouldn’t that be nice?

  13. Gravatar of Michael Rulle Michael Rulle
    4. June 2019 at 02:51

    If the market believed the BOJ was credible as defined then the policy would work. What about in the US where we currently price forward markets as if the FED were “anti-credible”—-or do we we believe they are credible enough to react to markets?

    In other words, when do rational expectations wag the dog? Is it rational to go against the (current) stated policy of the Fed?

    Can Central Bank be credible by not being credible (I.e., a belief that Fed will react to markets)?

  14. Gravatar of Todd Kreider Todd Kreider
    4. June 2019 at 05:34

    dtoh wrote:

    “7. Japan is doomed to a slow decline for at least the next two to three decades.”

    Why wouldn’t Japan also experience the coming A.I. and health/medical revolutions from 2020 to 2040?

    Since 2000, Japan’s per capita growth has been about the same as the U.S. and the U.K. and that will almost surely continue for the next 20 years.

  15. Gravatar of Michael Rulle Michael Rulle
    4. June 2019 at 13:00

    Today the market wagged the dog

  16. Gravatar of dtoh dtoh
    4. June 2019 at 16:15


    It depends a little bit on what you view as trend growth in the U.S., how you view the impact of tax cuts, and whether you buy in to Tyler Cohen’s secular stagnation thesis, but IMHO

    1. US per capita trend growth is at least double that of Japan. (In my view 2018 is pretty representative of trend growth going forward.)

    2. Demographics will exert increasing downward pressure on per capital output in Japan.

    3. AI is over-hyped and will achieve slow adoption in Japan.

    4. Healthcare is highly labor intensive and will be one of the last fields to benefit from AI. MHLW will continue, as they do now, to try to smother the introduction of new medical advances.

  17. Gravatar of Todd Kreider Todd Kreider
    6. June 2019 at 05:51


    1. Japan’s GDP per capita growth since 2010 has been 1.4%, about the same as 1.5% for the U.S.

    2. No.

    3. There is nothing over-hyped about current A.I. and where it will be in five years nor is there any reason to think Japan won’t adapt the latest technology as quickly as the U.S.

    Japanese used the internet at the same rate as Americans once it went to broadband and wasn’t that far behind earlier. Smart phone use lagged behind about a year.

    4. Healthcare won’t be labor intensive by 2029…

  18. Gravatar of H_WASSHOI (Maekawa Miku-nyan lover) H_WASSHOI (Maekawa Miku-nyan lover)
    6. June 2019 at 09:43

    *sigh* The next elections will be held on July.

  19. Gravatar of Greg Greg
    7. June 2019 at 02:54

    An metaphor involving expectations:
    If a government credibly commits to catching and severely punishing all criminals, observed rates of both crime and punishment will be low. Despite the positive correlation it is correct to say that punishment “works” to control crime. But it works via the expectations (of punishment) channel, rather than the actual punishment channel.

  20. Gravatar of Nick Nick
    13. June 2019 at 05:30

    Scott, what do you think would be a practical and credible policy that a central bank or government could implement? Buying about 100% of GDP of bonds has not obviously done a lot in the case of Japan, though it could be argued they were very low yielding liquid instruments anyway and as such not vastly different from cash.
    The central bank could give tax rebates linked to an inflation target financed by ‘printing’ money?
    the central bank could buy foreign financial assets or foreign or domestic real assets by ‘printing’ money? (I suspect anything foreign for a large economy would create all kinds of political fallout abroad)
    I’m Interested if you have some preferred way of creating money that would be most likely in your view to be implemented.

  21. Gravatar of ssumner ssumner
    13. June 2019 at 09:53

    Nick, I’d just go with buying government debt. The problems in Japan have nothing to do with what they are buying, it’s a lack of a “whatever it takes” policy that is the root of the problem. Japan’s policy is effectively contractionary, albeit slightly less so than before Abe.

  22. Gravatar of Nick Nick
    14. June 2019 at 00:16

    Thanks Scott, I am in agreement with you that lacking a whatever it takes credibility is a large part of the problem, i guess part of my question was “what would persuade people you really would do whatever it takes”. the ECB said we will do whatever it takes and no one really believes that.
    If Japan just continues what it is doing, in a not so crazy amount of time (certainly less than the maturity of many of their debt instruments) it will have no government debt not owned by the central bank and still << 1% inflation, maybe i am wrong to think that a tipping point wont occur and that it will just occur much later than expected?

  23. Gravatar of Robb Robb
    14. June 2019 at 07:04

    I agree with you that the Bank of Japan needs to have credibilty (hey it would be nice if our government was credible!), But I think it will take more than just buying more government debt.

    I think it would take some sort of “wild” statement/action to destroy years of disbelief. I think James Hamilton said it back in the day. I can’t find the exact quote but it was something like “buy up all the debt, buy up all the stockmarket, send everyone a check for a million dollars…somewhere in there we are gonna have some inflation”.

    I think that the situation has a whole lot more inertia than you seem to think.

Leave a Reply