Paul Krugman: Japan on the Brink: "Japan's current plan to hike consumption taxes a second time…
…has become… a sort of Rubicon for policy. And let me admit that people I respect-like Adam Posen, and some officials at international organizations-believe that Abe should go through with the hike. But I strongly disagree…. Right now, Japan is struggling to escape from a deflationary trap; it desperately needs to convince the private sector that from here on out prices will rise…. The pro-tax-hike side worries that if Japan doesn't go through with the increase, it will lose fiscal credibility and… the bond vigilantes will attack. Why don't I share that view? Partly because I don't see how this supposed crisis of confidence is supposed to work…. When a country borrows in its own currency and doesn't face inflationary pressure (quite the contrary), it's very hard to see how a Greek-style crisis is even possible. Short-term interest rates are controlled by the Bank of Japan; long-term rates mainly reflect expected short rates. Yes, investors could push the yen down, but that would be a good thing from Japan's point of view. Posen says stocks could crash, but I guess I don't see why if interest rates stay low and corporate Japan becomes more competitive thanks to a weaker yen. Seriously: tell me how this is supposed to work… [how a] fear that Japan might eventually monetize some of its debt-isn't actually a positive development. Meanwhile, it seems to me that Japan should be very, very afraid of losing momentum in the fight against deflation…. Could I be wrong?… Of course…. But it's all about weighing the risks. Right now, the risk of losing anti-deflation credibility looks much worse than the risk of losing fiscal credibility. Please, don't hike those taxes!
Continuing to worry my head about our intellectual adversaries here - including the very sharp and serious Adam Posen - over their fear that unless Japan raises taxes to begin closing its (admittedly huge) current budget deficits it runs serious risks of becoming "Argentina".
I get that part of the argument is that Japan can hit the same nominal GDP growth target by pairing a looser monetary with a tighter fiscal policy, and should do so. And to the extent that that is what is at issue - a call for fiscal tightening coupled with even more aggressive monetary loosening to hit the same nominal GDP growth target, and that what is being advocated is not just an increase in taxes but an increase in taxes coupled with full monetary offset in the form of additional monetary goosing, I get the argument. I even agree with it.
But to the extent that it is more than that…
The basic model of the taxmongers is, I think, the following:
- E(π) = π + δ(rD - σ) :: Inflation Expectations
- π = E(π) + β(u* - u) :: Phillips Curve
- r = r* + γ(u - u) + θ(π - π) :: Monetary Policy
Inflation Expectations: Expected inflation E(π) is equal to current inflation π plus some parameter times the difference between debt amortization rD and the expected primary surplus of the government σ.
Phillips Curve: Current inflation π is equal to expected inflation E(π) + a parameter times the difference between the natural rate of unemployment and the actual rate of unemployment.
Monetary Policy: The higher either the gap between the current inflation rate π and the central bank's target inflation rate π* or the higher the required gap between the natural rate of unemployment u* and the current unemployment rate u, the more the central bank must raise the interest rate r over the Wicksellian natural rate r* in order to achieve its monetary policy target-and then the higher is required debt amortization rD.
It then follows that the unemployment rate and the real interest rate will both be increasing functions of the fiscal financing gap rD-σ:
- u = u* + (δ/β)(rD - σ)
- r = r* + (γδ/β)(rD - σ) + θ(π - π*)
Taking differentials in response to a shock dr* to the Wicksellian natural rate r*, we get:
- du = D(δ/β)dr
- dr = dr* + D(γδ/β)dr
- du = [(δ/β)D/(1 - D(γδ/β))]dr*
Which tells us that if the debt D grows so large that Dγδ/β approaches one, even a very small adverse shock to the Wicksellian natural rate of interest dr* could cause the unemployment rate u to explode - unless the central bank abandons its monetary policy of inflation control - that is, of non-permanent-monetization of the debt.
For a country that does not borrow in its own currency, it is very easy to see why it must seek to avoid even a whisper of debt monetization. Such a whisper is an upward shock to E(π), and to the extent that is transmitted through to the current inflation rate such transmission produces an immediate jump in required debt service which makes the situation much worse.
But in a country that does borrow in its own currency and does control its own interest rates debt monetization and a resulting burst of inflation is no biggie: some of the debt is no-longer interest-bearing D that must be amortized but is money M. And to the extent that the rest of the debt has a duration greater than zero the increase in the price level reduces the value of the debt and thus the seriousness of the debt overhang problem.
Yes: I realize that this is arcana imperii. I do realize that I am not supposed to point out that reserve-currency issuing sovereigns with exorbitant privilege that thus control their own interest rates and borrow in their own currencies have a degree of freedom to use inflation as a tool of debt management that the Argentinas of the world do not. But an upward shift in expected inflation is what we are trying to generate here and now in Japan. And such an upward shift is only to be feared if Japan pretends that it is Argentina, and that it thus has no ability to monetize any of its debt.
If you are Argentina, then yes, sure: as Dγδ/β approaches one you get into the territory where a small upward shock to interest rates will cause either a Great Depression or force a price-spiral that, absent a currency reform, turns into hyperinflation.
But who thinks that Japan is Argentina?