The MMT position on hyperinflation is that it is sparked by an initial sharp contraction in output and the supply potential of the economy. (Bill Mitchell provides an in-depth explanation in this post.) For example, Mitchell points out that in Weimar Germany, the French and Belgian armies retaliated to the German default on reparations payments by taking control of the industrial area of the Ruhr. German workers stopped work in response. This resulted in a massive contraction of supply relative to demand.
In Zimbabwe, a case involving a fiat-currency issuing government which Mitchell discusses in some detail, the initial massive contraction in supply occurred as a result of the black backlash against an unjust distribution of resources. Land was confiscated and taken over by people not experienced in putting it to productive use. The motivation for the land reforms is understandable, but the immediate economic effect was a dramatic contraction in supply, with the country’s food output capacity almost cut in half.
Critics of expansionary monetary and fiscal policy are often quick to point out that hyperinflation is always accompanied by a massive expansion of the money supply (defined as currency plus demand deposits). This is true, but does not negate the MMT argument. Consider the quantity equation, which is an identity true by definition:
MV = PY
where M and V are the quantity and income velocity of money in circulation, P is the general price level and Y is real output.
In the case of Zimbabwe (and Weimar Germany), there was an initial sharp contraction in supply, and hence real output Y. Let’s say real output fell from Y to Y1 (Y > Y1):
MV = PY1
In such a situation, if there is no immediate change in MV on the left-hand side of the identity, there must be a massive increase in the general price level, P, to conform to the identity. So for any given money supply P must be much higher than before. The initial impetus for the hyperinflation has been a massive decrease in real output, Y, which even for given MV requires a massive increase in P.
According to Mitchell in the link provided, the reduction in food supply combined with a severe neglect of infrastructure (e.g. in industrial rail transport) amounted to a 60 percent contraction in potential supply. There is no easy way out of such a mess. If the government had attempted to contract the money supply, M, by 60 percent to keep prices from rising (or in an attempt to reverse their rise), this probably would have resulted in even more extreme poverty and starvation than has occurred. Continuing to add government spending, which creates net financial assets (defined as currency, reserves and treasury bills), doesn’t help much either because nominal demand (PY) has already hit the productive limits of the economy. Since Y cannot be increased further until the supply-side problems have been addressed, further increases in MV can only be resolved through a higher price level, P. This does not contradict MMT, though, which makes clear that government expenditure faces real resource (i.e. supply-side) limits. By the same token, there are political reasons why the government continues to spend in the face of a massive supply-side contraction. The resulting inflation redistributes income to debtors at the expense of lenders and savers (who, in the case of Zimbabwe, primarily comprise a wealthy 1 percent of the population that prior to the land confiscations owned 70 percent of the productive land and a high proportion of the country’s wealth). Without this redistribution, matters probably would have been even worse for the general population.
So, from an MMT perspective, a common factor in episodes of hyperinflation is an initial sharp contraction in supply. In contrast, the normal situation of a capitalist economy is one of excess capacity and demand falling short of potential supply. This is even more true during economic downturns such as the present one. In these circumstances, an increase in MV will not necessarily be associated with inflation because there is plenty of room for real output, Y, to be expanded in response to stronger demand (whether coming from private, public, or external sources). An increase in MV can correspond to an increase in PY that is primarily or even entirely due to an increase in Y. There is a competitive pressure on firms to respond to higher demand through expansions of output rather than increasing prices whenever this is possible. Firms that fail to do so risk losing market share to competitors.
Broad money, M, is endogenously determined, largely through the lending behavior of private banks. When economic activity strengthens, the prospects for productive investment improve and the capacity of households to service debt also improves, and this increases the willingness of private banks to lend. In lending, the banks simultaneously create new deposits (new M), which results in an increase in MV (V is likely to be rising as well during an expansionary phase). Causation, in contrast to the orthodox view, runs from PY to MV. It is the pick-up in economic activity, PY, that increases the willingness of private banks to lend and expand the effective money supply, MV. Banks do not require prior reserves to extend these loans. They can always access the necessary reserves after the fact. In particular, the central bank is committed to supply reserves on demand. An implication is that the central bank is incapable of exogenously controlling M. It can control the short-term interest rate, and in fact can control the entire yield curve if it wants to, but by necessity must allow M to vary in response to the lending behavior of private banks.
As long as there is still scope for real output to be expanded, the expansion of the effective money supply brought about through increased bank lending facilitates an increase in real output. The nearer the economy gets to full capacity, the more likely it becomes that further expansions of MV will be accommodating increases in P rather than Y. It is only at that point that MMT suggests the government should be cutting back its net expenditure to prevent overheating.