Inflation is always and everywhere a political phenomenon
don't understand inflation. Those who lived through the high
inflation of the 1970s are convinced that inflation is always and
everywhere caused by wage-price spirals. Germans, and economic
Austrians, are convinced that inflation is always and everywhere
caused by central bank money printing. Small-state supporters are
convinced that inflation is always and everywhere caused by
profligate governments borrowing and spending excessively. Hard money
enthusiasts are convinced that inflation is always and everywhere
caused by currency devaluation. Every economic school has its own
theory of inflation. MMT, which until recently had no clear theory of
inflation, was roundly criticised for this by mainstream and
heterodox economists alike: it is now developing a theory which as
far as I can see is based upon the idea that inflation is always and
everywhere caused by inefficient deployment of labour. Hmm.
In fact we don't even know what we mean by inflation. As the Cleveland Fed entertainingly discusses, inflation originally meant expansion of (paper) currency in a manner that resulted in higher prices. But over time, that definition has widened to mean anything and everything that raises prices, not just monetary expansion. And not only consumer prices, either. We now talk about inflation in asset prices, profits and wages as well. Keynes could be blamed for this – he talked about different “types” of inflation. But it is probably fair to say that he was simply tapping into commonly-held beliefs about inflation in his time.
Measuring inflation is something of a black art. Central banks use several different measures of price inflation. Officially they target consumer price indexes which are based on baskets of typical consumer goods. But there may be several versions of these using different methodologies and giving different results.
And the baskets themselves are problematic. Some of the items in the basket may have volatile prices, which can cause inflation to appear more unstable than it really is. And as people's shopping habits change over time, the basket has to be updated periodically, which can cause sudden spikes or dips in inflation measures. Some goods are subject to technological improvements which mean that their real price falls over time even though the nominal price is unchanged: other goods are subject to marketing “improvements” that mean their real price rises even though there is no change in the nominal price. Statistical agencies do attempt to adjust for such changes – this is known as “hedonic adjustment”. But this is a complex and difficult task, and it is questionable how accurate the results are anyway. In anything more developed than a pure commodity market, price rises are more a matter of opinion than evidence.
But central banks also use other measures of inflation that don't tend to hit the headlines. “Core” inflation, which excludes goods that have volatile prices or are subject to external shocks, is the preferred measure for central banks that target medium-term inflation. The problem with this is that the goods typically excluded include food, energy and fuel, which are major expenditure items for ordinary people. When “core” inflation deviates significantly from CPI, for example after an oil price shock, it can lead to accusations that the central bank is not really targeting consumer price inflation at all.
The other measure of inflation used by central banks is the implied GDP deflator, which is usually expressed as the ratio of nominal and real GDP. Nominal GDP is determined from the current year's prices, while real GDP is calculated by reference to a base year. But the calculation of both nominal and real GDP is itself a complete can of worms which I won't go into here: for those who are interested, I recommend Diane Coyle's excellent book on the history of GDP. Suffice it to say that the implied deflator, while perhaps the most reliable measure of inflation because it does not rely on a fixed basket of goods, is only as good as the measures of GDP from which it is derived. Hmm.
None of these measures of inflation include asset prices, wages or profits. Central banks have been roundly criticised for failing to prevent the large rises in house prices that led to the financial crisis. But asset price rises were then, and still remain, outside central banks' inflation targeting mandate, and wage and profit rises are only within it to the extent that they feed through to consumer price inflation. They may however form part of a central bank's financial stability mandate.
So we don't understand the causes of inflation, we don't agree about what we mean by inflation and we have no reliable means of measuring it. Yet we are absolutely terrified of it. And we want government (via its central bank) to make sure that inflation NEVER HAPPENS. How very dare government rob us of our precious savings by means of inflation!
Underlying this statement, and indeed all statements about the control of inflation, is a powerful and fundamentally irrational belief. Inflation can be prevented by government. Therefore, if inflation happens, it is because government has allowed it to. Inflation is therefore always and everywhere a POLITICAL phenomenon.
This belief may be irrational, but that doesn't necessarily make it wrong. Do governments really cause inflation? Can they really prevent it?
Researchers at the Dallas Fed think they do and they can. In this paper, they endorse what is known as the “fiscal theory of the price level”. Basically, a government that can credibly guarantee the future purchasing power of the currency can prevent inflation taking hold. This is essentially the same as the monetarist argument that a central bank that can credibly commit to controlling the money supply can always prevent inflation taking hold. Monetarists and fiscalists fight to the death about whether fiscal or monetary policy is better at controlling inflation. But this is a completely redundant argument. As far as inflation control is concerned, it does not matter whether the controlling authority is government or central bank, or the policy being pursued is fiscal or monetary. What matters is that the authority is CREDIBLE.
To see how important credibility is, it is instructive to look at cases of hyperinflation. Hyperinflation occurs when people reject a currency. They dump their holdings of it as fast as they can and they refuse to use it as a medium of exchange. A currency that people won't use is by definition worthless. Hyperinflation is the destruction of a currency. The question is, what causes people to reject a currency?
It is widely believed that hyperinflation is caused by central banks printing money. The quantity theory of money says that when the quantity of money increases more than production, the price level increases. Giles Wilkes describes how hyperinflation might occur in a gold standard economy if there is a sudden massive inflow of gold. In a fiat money economy, the equivalent would be a central bank suddenly creating and distributing very large amounts of currency. So out-of-control money printing by irresponsible central banks, especially to monetize the profligate borrowing of corrupt governments, must cause hyperinflation – mustn't it?
Not necessarily. The Dallas Fed research shows that hyperinflation can happen even when there is no increase in the monetary base, if people lose faith in the ability of government to guarantee the future purchasing power of the money already in circulation. Therefore money printing by central banks cannot possibly be the primary cause of hyperinflation – though the Dallas Fed researchers do say in a footnote that direct monetization of government debt issuance would be inflationary.
But note the adjectives typically used about governments in countries experiencing monetary hyperinflation. I've used some of them above: “Irresponsible”. “Profligate”. “Out-of-control”. “Corrupt”. Would irresponsible, profligate, out-of-control and corrupt monetary and fiscal authorities be credible? Hardly. They would be basket cases. Monetary financing of corrupt and incompetent government by a captive and irresponsible central bank is indeed inflationary. The lunatics are in charge of the asylum, and everyone knows it. Of course people reject the currency.
The most recent example of monetary hyperinflation is Zimbabwe, where a corrupt and incompetent government trashed its main industry then printed money to pay its creditors and cronies. The result was hyperinflation as the people of the country, suffering a catastrophic loss of confidence in the government, rejected the currency. The Zimbabwean dollar is still not accepted in Zimbabwe. Lots of other currencies are though.....after all, if you lose confidence in your own government you turn to others with a better track record, don't you?
Anchoring to something that people can trust is the usual method of ending hyperinflationary episodes. The Dallas Fed observes that the Weimar hyperinflation ended when a new currency, the Rentenmark, was created which was anchored to real estate. Restoring a gold standard would have been an alternative. But by far the most common method is pegging to a more stable currency, or even adopting it outright as Zimbabwe did. The currency most frequently used is the US dollar, but other currencies that may be used include sterling, yen and Swiss francs. In Eastern Europe, still scarred by hyperinflationary episodes after the fall of the Iron Curtain, the currency of choice was formerly the German Deutschmark and is now the Euro.
The most extreme form of loss of confidence in government is war, particularly for those on the losing side. Hyperinflation is common after wars: there were clusters of hyperinflations after all three of the major wars of the 20th Century (World War 1, World War 2 and the Cold War). The worst hyperinflations have been in Eastern Europe, which has a long and recent history of war and political instability, and China, which is historically among the most war-torn and politically unstable countries in the world.
Really, the root cause of hyperinflation is social and political chaos. Money printing and monetization are symptoms of hyperinflation, not causes: they are only hyperinflationary when there is no credible fiscal or monetary authority. But hyperinflation is exceptional. What of more ordinary inflation?
The Dallas Fed researchers regard hyperinflation as simply extreme inflation. They therefore apply essentially the same causation to ordinary inflation. For them, inflation is a consequence of lack of fiscal & monetary credibility. A credible government can prevent inflation completely if it chooses.
The Dallas Fed researchers think government credibility comes from commitment to future primary surpluses. But this doesn't make sense: no
democratic government can credibly commit to delivering primary
surpluses beyond its electoral span. Personally I think that a history of generally honouring obligations
– and in the case of the US, a constitutional commitment to honour
obligations - is far more important than any non-binding (and
impossible) “commitment” to future primary surpluses. The world
simply expects the US to do whatever is necessary to meet future
obligations, whether that be raising taxes, selling assets, ponzi
borrowing or even – horrors – money printing. There is probably
an element of what we might term enlightened self-interest, too:
no-one wants the US dollar to collapse, so the world is likely to
make sure that it doesn't. Expectations can be founded on wishful
thinking, and wishes can come true.
The Dallas Fed researchers say it is the stance of the fiscal authority, not the monetary authority, that matters for the control of inflation. This is not unreasonable: fiscal dominance, where the government makes irresponsible spending commitments that force the central bank to monetize debt ex post, is known to be inflationary. And we now know that fiscal dominance can apply on the downside too: a government hell-bent on fiscal austerity in the teeth of a recession can seriously hamper the monetary authority's attempts to generate recovery, as we have seen in recent years. But it simply isn't true that a government that runs persistent deficits is necessarily encouraging inflation. It depends on the country and the circumstances. Venezuela running persistent deficits is likely to experience high inflation because of its unstable political situation and its history of fiscal irresponsibility. The US, with a stable democratic government and a history of meeting its sovereign obligations*, is much less likely to.
And this brings me back to the title of this post. If the Dallas Fed researchers are correct that it is the fiscal authority's stance that determines the path of inflation, and the fiscal authority is subject to election by popular vote every few years, then inflation is by definition a political phenomenon.
But even if the Dallas Fed paper is wrong, and it is central banks that really rule the roost, inflation is still a political phenomenon. The independence of central banks is an illusion. Central banks are only as independent as politicians allow them to be. As long as politicians want inflation to be low, central banks will be required to keep it low, even if they would rather allow it to rise to encourage spending and growth. And as long as the popular vote is dominated by people who want inflation to be low, politicians will want to keep it low. So if you want inflation to rise, kill the old and the rich. They are the people who have most to lose from higher inflation – and they vote.
But be careful what you ask for. It is the influence of inflation hawks on government and central bank that gives them the credibility to control inflation. And - unsurprisingly - a high proportion of inflation hawks are old and/or rich. Without the old and the rich, would inflation targeting be possible at all?
Hyperinflation – It's more than just a monetary phenomenon – Cullen Roche
The (political) failure of safe assets - Coppola Comment
Central banks, safe assets and that independence question - Coppola Comment
* To be fair, the French think otherwise. They still regard both the Nixon shock and FDR's suspension of the gold standard as sovereign default.
Image from Noah Smith. The related post is well worth reading.
Keep up to date with the latest thinking on some of the day's biggest issues and get instant access to our members-only features, such as the News Dashboard, Reading List, Bookshelf & Newsletter. It's completely free.