The sky’s the limit: The economics of inflation and hyperinflation

6 Money printing and inflation

To learn more about the origins of the global financial crisis, read Sections 8.1 and 8.8 of The Economy 2.0: Macroeconomics.

To learn more about quantitative easing, read Section 6.7 of The Economy 2.0: Macroeconomics.

To learn more about the central bank’s role in the financial sector and how it interacts with commercial banks, read Section 6.6 of The Economy 2.0: Macroeconomics.

quantitative easing (QE)

Central bank purchases of financial assets aimed at reducing interest rates on those assets when conventional monetary policy is ineffective because the policy interest rate is at the zero lower bound. See also: zero lower bound.

base money, monetary base, high-powered money

Base money (also called the monetary base and sometimes high-powered money) consists of the cash held by households, firms, and banks, together with the balances held by commercial banks in their reserve accounts at the central bank.

In 2007 a mortgage crisis in the US led to a banking crisis and a global financial crash. In response, central banks around the world slashed interest rates, and some embarked on major purchases of financial assets in the operation known as quantitative easing. To purchase these assets they printed new base money—which is to say, they paid for the assets by crediting money to the commercial banks in those banks’ accounts in the central bank, creating new base money. In the US this led to the monetary base more than doubling from 2007 to 2009, and more than doubling again from 2009 to 2014. Figure 7 shows this explosive rise in the monetary base. It also shows the inflation rate, which averaged just 2.0% over 2007 to 2014, lower than before 2007.

The Nobel prize-winning economist Milton Friedman once wrote that ‘Inflation is always and everywhere a monetary phenomenon in the sense that it is and can be produced only by a more rapid increase in the quantity of money than in output.’ In this section, we will show why printing money is sometimes an important part of the story in cases of very high inflation. Nonetheless, the quote is mistaken in two ways. First, our models above show that we do not need to appeal to the quantity of money in order to explain inflation: inflation can be produced by a wage–price spiral, a decline in the terms of trade, or a depreciation–inflation spiral without any reference to the quantity of money. But second, and perhaps more surprisingly, the experience since 2007 shows that printing enormous quantities of money may have little or no effect on inflation. What explains this outcome?

The answer to this last question is that the new base money did not push up prices of goods and services, simply because it was not spent on goods and services. Quantitative easing did indeed push up the prices of the financial assets that the new money was used to purchase, and deliberately so: the hope of the central bank was that the positive wealth effect of rising asset prices would lead the private sector to spend more, stimulating the economy. But the impact on prices of goods and services was very small at best, as shown by the continuing low rates of inflation. (Since the central bank was concerned about falling into deflation, which was successfully avoided, they might argue that the policy was a success.)

To learn more about the difference between bank money and base money, read Section 6.8 of The Economy 2.0: Macroeconomics.

bank money

Money in the form of deposits in commercial banks. The bank allows bank deposits, created for example when the bank makes a loan, to be used as a means of exchange, debiting the buyer’s deposit and crediting the seller with a new deposit.

Moreover, the lack of a clear link between printing money and inflation shouldn’t be surprising when we remember that what governments print is base money, whereas most of what people spend is commercial bank money. Bank money does not depend directly or in any simple way on the quantity of base money—it arises from bank lending, which depends on the policy interest rate (the ‘base rate’) and economic conditions. Bank money is the sum of everyone’s bank balances, and this is very much larger than the amount of base money that the central bank has produced. So in the case of a monetary expansion (loosening of monetary policy), the usual process is the following:

1. reduce the policy interest rate, in order to
2. stimulate desired private sector spending, which leads to
3. more loans being taken out, which implies
4. creation of commercial bank money.

To learn more about the zero lower bound, read Section 5.9 of The Economy 2.0: Macroeconomics.

After the global financial crisis many countries (mainly the high-income countries) cut interest rates close to zero. Further cuts are not possible at the zero lower bound, which is why they resorted to quantitative easing. And this demonstrated that governments printing money might not in itself have a significant effect on prices of goods and services.

Monetary deficit financing

In some circumstances, however, printing money really does make a difference.

default

A borrower who fails to repay a loan, or repays less than is required under the contract, is said to default on the loan. More generally, any failure to meet the terms of a contract can be described as a default.

Governments are required to print money when they have a fiscal deficit that they cannot cover by borrowing. This occurs when the private sector is unwilling to buy government bonds in sufficient quantity. It is likely when investors don’t trust that the government will pay back what they borrow, for example when investors fear that the government might default on its bonds. Or it might happen if investors think that the real interest rate is too low compared to that of other countries, which can happen because the nominal interest rate is too low relative to expected inflation.

A fiscal deficit that is financed by printing money is likely to be more inflationary than a deficit financed by selling bonds for the following reason: When a deficit is financed by borrowing, every peso (for instance) that the private sector uses to buy a government bond is a peso less of private expenditure, or aggregate demand. In this case, government borrowing for expenditure is the mirror image of private sector saving, and that saving partially offsets the government expenditure. But when the government runs a fiscal deficit that is financed by printing money, there is no such offset. Demand is increased by government expenditures without being reduced by either taxes or private-sector borrowing. So aggregate demand increases further, leading to a larger bargaining gap and inflationary pressure in the WS–PS model.

While printing money to cover a fiscal deficit tends to cause inflation, there is also causality in the opposite direction: inflation causes the government to print more money. Why is this? Consider an independent central bank with an inflation target. The central bank chooses the interest rate it thinks will achieve its inflation target, and supplies whatever quantity of base money is required in the money market to achieve that interest rate on government bonds.

But the amount of base money required to sustain that interest rate depends on the amount of money that banks need for their daily transactions. This in turn depends on nominal GDP (the nominal value of transactions in the economy). That nominal value rises with inflation. So if nominal GDP rises by 10% then, ceteris paribus, commercial banks will demand 10% more base money at any given policy interest rate. This requires the central bank to create 10% more base money. This means that the quantity of base money is endogenous: it is the outcome of a combination of economic policy and economic conditions.

A government is not forced to print money in this scenario. It always has the option simply to stop supplying more money to the money market. But the effect would be a spike in the interest rate: higher demand for base money without higher supply will push up the price of borrowing. This means that the central bank cannot choose a particular policy interest rate and halt growth in the supply of money.

So causality runs in both directions: from money printing to inflation and from inflation to money printing. And since high and persistent inflation is often caused by the depreciation–inflation spiral (discussed in Section 4), depreciations also tend to lead to money creation. Historically, during episodes of very high inflation, money growth has followed pre-existing episodes of inflation or exchange rate depreciation more often than it preceded them. This suggests that money growth is more often an endogenous effect of other economic dynamics than it is a driving force.¹

Seigniorage and the inflation tax

Still, sometimes governments have an independent motive to print money, because money printing represents revenue to the government: it is spending over and above its tax revenue that does not increase its debt. Like the wage–price spiral it is also the result of conflicting claims over output, but instead of workers and firms, it arises out of the conflict between taxpayers on the one hand, and the recipients of government expenditure on the other, including public employees, pensioners, and the recipients of subsidies. The deficit is a result of a failure to negotiate higher taxes with taxpayers and/or lower expenditures with recipients of government expenditure. This is particularly likely to occur when a government is trying to boost support before an election. And when this occurs in a context of low demand for government bonds as discussed earlier—either because they are perceived as risky, or because the real interest rate is unattractively low or negative—then the outcome is printing money in order to try to satisfy everyone.

The sociologist Fernando Henrique Cardoso was appointed finance minister in Brazil in May 1993, when the monthly inflation rate was 27%. After implementing a stabilization plan called the Real Plan (which we will discuss further in Section 7), the inflation rate fell to just 2% per month in August 1994. In October that year he was elected president of Brazil, serving two terms. For Cardoso, inflation reflected a failure to resolve the multiple conflicting demands faced by the government. As he describes in his memoir, ‘As long as there was inflation, [politicians] never had to say no to anybody. No spending request was too large; just print more money! It was a diabolically efficient way to solve disputes. Even though inflation would eventually take away any gains, at least everybody felt good in the short term.’²

While this is an important observation, Section 7 will show that there is much more to the story: this kind of conflict has led to extreme inflation only in the context of a major negative shock such as a debt crisis. The conflict is not so much over gains from government policy as it is due to each group trying to minimize its own losses in the face of a decline in overall living standards.

seigniorage

The revenue the government gains when the central bank creates new base money.

The revenue that money printing implies for the government is called seigniorage. Governments typically receive some seigniorage revenue automatically as nominal GDP rises. This is because higher nominal GDP implies higher demand for base money at any given interest rate, and the government (in this case the central bank) will supply this extra money. In most cases, with low levels of inflation, this amounts to very little real income for the government. But when inflation gets extremely high, seigniorage can take on more significant proportions, as illustrated for some South American countries in Figure 8. Bolivia’s peak of 15.8% of GDP in 1984 coincided with inflation hitting 1,281% that year.

If the change in the supply of base money is \(\Delta M\), then the real revenue received by the government is

\[\text{seigniorage} = \frac{\mathrm{\Delta}M}{P}\]

where \(P\) is the price level.

For non-index-linked bonds, high inflation reduces the real interest rate (and it could become strongly negative), in which case we might also include the reduction in real government debt as part of seigniorage. But since we cannot know what the real interest rate would be in a counterfactual world with lower inflation, any quantitative estimate of this value would be highly speculative. So we will ignore it here, but note that ‘inflating away’ debt can be important in some cases.

We can rewrite the seigniorage equation as

\[\text{seigniorage} = \frac{\mathrm{\Delta}M}{M} \times \frac{M}{P}\]

So seigniorage is the growth rate of base money times the real quantity of base money in the economy.

This equation shows that seigniorage is rising in the amount of money printed but declining in \(P\). This means that as prices rise, more money must be printed to maintain the same level of real seigniorage. This will become important when we discuss hyperinflation in Section 7.

For governments that are struggling to either reduce their fiscal deficit or to cover it using debt, printing money is a very tempting option for raising resources. For this reason, high levels of money printing are typically associated with major crises such as revolutions or civil wars which drastically reduce a government’s capacity to raise taxes, or with debt crises which drastically increase government’s burden of debt repayment. Such crises not only cause a fiscal deficit, but they also make it very hard for the government to issue new debt, because the private sector will be concerned that the government will default.

inflation tax

The implicit transfer of resources from the public to the government caused by inflation eroding the value of money holdings.

Since printing money does not create new real resources, seigniorage revenue must come at the expense of other economic actors. This occurs via the inflation tax: the rise in prices implies that people’s existing holdings of money decline in real terms. The real value they lose is effectively transferred to the government (as discussed in the ‘Find out more’ box).