Paul Krugman has been dismissing concerns about bond vigilantes. From the point view of a Market Monetarist, bond vigilantes are bad, but there is a silver lining to their cloud. And it is that silver lining which provides the element of truth to Krugman's argument.
I think Krugman's framing is that "other people" are claiming that it is necessary to raise taxes and cut government spending soon in order to reduce the budget deficit. Yes, this will tend to slow or reduce spending on output and so slow the already weak recovery or force the economy back into recession. But, according to some, if we don't deal with the budget deficit soon, the bond vigilantes will strike. They will sell off bonds and force up interest rates. The higher interest rates will tend to slow or reduce spending on output and so weaken the weak recovery further or return the economy to recession.
I think Krugman is arguing that this is wrong, and that when the bond vigilantes strike, they force down the value of the dollar, and so increase spending on domestic output--more spending on exports and more spending on import competing goods. Fiscal austerity, then, will tend to depress spending on output, while an attack of the bond vigilantes will have the opposite effect. The notion that we are going to have slower spending on output anyway, so we might as well reduce the budget deficit, is wrong.
From a Market Monetarist perspective, monetary policy can and should be used to keep nominal spending on output on target. With a competent monetary authority, bond vigilantes won't have any effect on total spending on output. Still, if people holding government bonds expect that the monetary regime will hold, but government will explicitly default, or else the regime will break down and money will be created to pay off the debt, or that the real exchange rate will later depreciate so that holding U.S. bonds is less attractive than foreign bonds, there can be an immediate sell off of bonds and depreciation of the U.S. currency.
The decrease in the value of the dollar results in higher prices of imported consumer goods and lower real incomes. The reason for the lower real incomes is the less advantageous terms of trade. To keep spending on output from rising above target, market interest rates must increase. Those fixated on interest rate targeting would frame this as the central bank raising its target interest rate to dampen increases in spending.
From a Market Monetarist perspective, selling off bonds lowers their prices and raises their yields. This does tend to reduce spending on all sorts of output, but this is just offsetting the increases in spending on import competing goods and export goods. If the higher interest rates result in people choosing to hold less money, then the monetary authority should reduce the quantity of money, keeping an excess supply of money from pushing spending on output above target.
As for the government's budget, it now has a greater interest expense. It must either raise taxes or else cut government outlays other than interest. Whatever benefits those government programs provided are sacrificed, while those paying higher taxes or receiving reduced transfers will enjoy fewer consumer goods and services either now or in the future. Since much of the U.S. debt is held externally, it is pretty clear that those domestic reductions in well being go to pay off foreign debt, which is the other side of the coin of the additional exports and reduced imports.
That is all bad.
Now, Market Monetarists believe that spending on output is currently too low, well below the growth path of the Great Moderation and below an appropriate start for a new regime. For example, I believe that a Reagan/Volcker nominal recovery is in order, before starting with slow steady nominal spending growth. If the bond vigilantes attacked, and spending on output was allowed to rise rapidly, then this would help generate an appropriately rapid nominal recovery. This is the element of truth in Krugman's perspective.
To the degree that real output and employment expand due to more rapid growth in spending, then this added real output (a closing on the output gap,) will partly or perhaps fully offset the adverse effects of the lower exchange rate and higher interest rates. To some degree, formerly unemployed resources would be used to produce the export goods that pay back the foreigners. The reductions in the transfers received and increase in the taxes paid by the formerly unemployed would help fund the increase in interest expense on the national debt.
But, of course, Market Monetarists (unlike Krugman,) believe that it is in the power of the Fed to bring about a rapid nominal recovery without an attack by the bond vigilantes. And so, there is really no particular benefit to such an attack. Further, Market Monetarists believe that the Fed can offset the effect of any fiscal austerity, and so, prefer that fiscal policy be aimed at efficiently funding an appropriately-sized government. It is the monetary authority's job to make sure that adjustments in fiscal policy, like the myriads of other things that might impact various elements of spending on output, remain consistent with the nominal anchor--ideally a target growth path for spending on output.
If, on the other hand, the Fed were to target the exchange rate, then an attack of the bond vigilantes would require a rapid contraction in spending on output. Further, a focus on inflation, at least of consumer goods, would also require a contraction of spending on output to prevent the inflation of the prices of imported consumer goods.
This second scenario is very realistic, and so, unless inflation is running below target, an attack by the bond vigilantes could be very contractionary. It seems to me that this is the framing that those favoring fiscal austerity have in mind when they worry about the threat of the bond vigilantes. For a Market Monetarist, inflation targeting is a bad idea, and exchange rate targeting is worse.