Too Fed to Fail

[Admiral Yi]

Hence the rage for independent central bank structures following the 70s.  But that has issues of its own.  One could see the present structure as a way to maximize technocratic control over the economy without causing the peasants to revolt.  There is something to be said for that.  Also something to be said against that

One could and should see it as a way to take control of monetary policy away from elected leaders who, time and again, and have demonstrated their propensity to buy votes in the present with the illusion of cost free affluence.  The only thing to be said for that is the illusion of cost free affluence.

  The government could still issue bond if it wished, even absent a budgetary function - see below

Or it could not.

War finance presents the same problem as now: whether the government can create a credible commitment to undo the monetary expansion once the crisis ends.  If the government;s commitment is credible then there won't be an expectation of permanent monetary enlargement and inflation can be kept under control.  Under this schema that could be done the traditional way - selling warbonds to mop up the excess liquidity with a commitment to retire the bonds in the future 

You misread my hypothetical.  Think Vietnam.  Easy money was used to finance the war, which led to 70's stagflation.  Now one could argue that revenues should have been raised to finance the war, but the case can also be made that wars are exceptional events [insert mongers-type joke here] which can and should be financed through deficit spending.  In this one handed system you can't run a deficit and maintain steady, non-inflationary money supply.

[The Minsky Moment]

You misread my hypothetical.  Think Vietnam.  Easy money was used to finance the war, which led to 70's stagflation.  Now one could argue that revenues should have been raised to finance the war, but the case can also be made that wars are exceptional events [insert mongers-type joke here] which can and should be financed through deficit spending.  In this one handed system you can't run a deficit and maintain steady, non-inflationary money supply.

I think Vietnam shows that having a two-handed system doesn't fix that problem.

In either system the problem is the same: in order to finance war without generating inflation, the government must make a credible commitment to run G-T<0* for a period of time after the war concludes.
In a two-hand system, if the government chooses to finance by selling bonds, it must convince investors that it will be able to repay those bonds without unduly inflating away their real value.  That means the future commitment to keep G-T down.  Otherwise either people won't buy the bonds or the debt burden will be inflated away ex post.

In a one-hand system, the problem is the same.  A temporary increase in the money supply will not raise inflationary expectations if the population truly believes it will be reversed in the future.  In the one hand system that means the government committing to keep G-T down after the war's end. 
To the extent it helps convince the population of the credibility of the commitment, and to temporarily mop up excess liquidity that finds its way into private hands during the war, the one-handed government can issue warbonds just as the two handed one does.  But that will only work if the population believes the government will redeem them after the war by increasing the rate of T instead of just rolling the printer.  Again, same as in the two-handed case.

* really not strictly zero, but below RGDP growth after accounting for interest burdens.

[Admiral Yi]

In either system the problem is the same: in order to finance war without generating inflation, the government must make a credible commitment to run G-T<0* for a period of time after the war concludes.

* really not strictly zero, but below RGDP growth after accounting for interest burdens.

This is not true.  Countries can pack on additional debt, up the elusive tipping point when investors start to flee, without causing inflation.  National debt skyrocketed in the last 6 years, and inflation is running around 2%.  Surely you're not going to claim that this is because investors know that this additional debt will be paid off with surpluses for a period of time after Teh Greatest Recession is over.

[The Minsky Moment]

This is not true.  Countries can pack on additional debt, up the elusive tipping point when investors start to flee, without causing inflation. 

In a dual system like ours, the question is whether people believe the debt will ultimately be repaid through taxation.  If the "tipping point" is reached and people don't believe the government can or will repay the real value of the debt, either there will be a default or the expectation that the debt will be inflated away (the only difference being distributional consequences).  If the latter, then inflationary expectations will build.

In the unitary system it is really no different.  The government makes the expenditures directly, but there is no permanent increase in the money supply if the government commits to retire the issuance in the future by taxing it away, or cutting future spending.  A purely emphemeral increase in the monetary supply shouldn't affect inflationary expectations.  And if the government is concerned about temporary excess liquidity it can always issue bonds, in which case the result is exactly the same as in the dual system.

In both cases, the government's task is the convince the public that the expansion of spending will be reversed in the future, and in both cases the consequences of failure are either a buildup in inflationary expectations or a default.

National debt skyrocketed in the last 6 years, and inflation is running around 2%.  Surely you're not going to claim that this is because investors know that this additional debt will be paid off with surpluses for a period of time after Teh Greatest Recession is over.

Inflation is actually running well under 2% by some measures . . .

In our dual system, the issuance of debt should have zero inflationary impact unless the public believes either: (1) that the government will elect to resolve the debt in the future in part by inflating it away, or (2) that the underlying spending will jump start the economy onto a more expansionary (nominal) path.  The lack of inflationary expectations indicates that neither belief predominates in the markets, and IMO for good reason.

Inflation is low because notwithstanding the Fed's pumping reserves into the banking system, the reserves are for the most part just sitting dormant in Fed accounts and the private financial system is not creating new money by lending.  Whether this is demand side (firms cutting investment plans) or supply side (banks being cautious and restricting credit) or both is hard to say but the result is plain enough.  The markets looks ahead and see stagnant nominal growth.