By Matthew Nolan*
After a long period discussing tax, we are now up to the very last tax article !
The last type of taxation we are going to discuss is inflation tax.
Now to do this we cannot just say “inflation is bad”.
In truth, we need to ask how policy actions related to inflation and monetary policy can function as a tax, and why (as a result) economists often tend to steer away from forms of inflation taxation and direct money financing of deficits when discussing optimal tax policies.
Inflation in the context we are describing is persistent growth in the general price level – think of it as the way the real buying power of money declines, relative to goods and services, through time.
In 1978, Stanley Fisher and Franco Modigilani stated the following:
There is no convincing account of the economic costs of inflation that justifies the typical belief -- of the economist and the layman -- that inflation poses a serious economic problem, relative to unemployment
To be fair Fisher and Modigilani then go on to discuss many of the costs that are involved in a fairly sophisticated way. When I studied undergraduate economics I couldn’t appreciate these costs, with appeals to “shoe leather costs” and “menu costs” from inflation sounding overstated. If the increase in inflation also increased nominal wages then what was the problem?
In the end, the argument always seemed to be a call that “any drop in unemployment from boosting inflation is temporary, while the costs of higher inflation are permanent”.
While this is all well and true, it came off as a very uncompelling argument – I was not convinced it mattered. So while this is the argument you’ll often hear, I intend to flesh the details out a little more here in a hopefully accessible way.
Eventually I was taught a different way to think about the issue of inflation – by using the framework we have discussed for tax, and as a result considering what specific policy choices about money financing would have on the underlying distribution of goods and services!
Monetary policy, inflation, unemployment, and co-ordination
Before we can clearly think about issues involving inflation, we need to briefly think about the macroeconomic situation more generally.
Previously the forms of taxation we’ve discussed, where there has been a focus on the long-run, and in the case of externality taxes where the markets of interest are small relative to the economy, focused on microeconomic arguments. Now “microeconomics” does not mean small, it can indeed be over an entire economy (the idea of general equilibrium) – and it offers the logic we would like to use to understand the issue as a tax.
But inflation and money are a macroeconomic phenomenon. The difference with “macroeconomics” is that it involves thinking about concepts that are aggregations of smaller components, but that we may not be able to easily break down. The key macroeconomic variables we think about are inflation, money supply, unemployment rates, and gross domestic product.
Microeconomic arguments underlie these macroeconomic ideas, but ultimately as macroeconomists we recognise that individuals are making choices relative to these macroeconomic variables (as well as other things) and their expectations about them.
In that context, government policy can help to co-ordinate expectations about all these different macroeconomic variables. Note: A neat way of looking through one of the channels this runs through has been discussed recently here and here.
The debates within economics have led to a broad agreement that targeting a rate of inflation, which is anticipated by private individuals, and ultimately realised by their choices and the choices of an independent central bank will serve this role. While unemployment and GDP may be influenced by factors outside of our control (changes in the skills firms require to make products, droughts, changes in technology), inflation and the associated predictable change in the price level can be set so private individuals can make choices with more certainty.
An independent central bank can credibly commit to doing this, and given the information available the central banks in Australia and New Zealand have done an incredible job at smoothing out movements in inflation and as a result preventing excessive drops in employment and economic activity.
As a result, we will take as given that we have an independent central bank directly dealing with this idea of co-ordination – something you may have heard called “demand management” or “flexible inflation targeting” by many people. Other types of rule following co-ordination policies are sometimes mentioned such as Nominal GDP targeting – however, for all intents and purposes any such rule based policy is based on helping with this overall co-ordination of expectations where practicable.
Given this, what happens if the government turns around and tells the central bank that they want to lift spending and they want it to be financed by central bank funds?
Unanticipated monetary stimulus
Assume that we are in a world where there is inflation targeting, such that the central bank has set a time path of interest rates that it view as, given current information, appropriate for the economy.
Then the government decides that it wants to temporarily increase spending, and it wants the central bank to fund this for them. In this context, the government is solely interested in increasing the goods and services they produce/transfer – but exactly how this is paid for will end up being determined by the central bank.
There are two ways the central bank could deal with this sudden claim:
1. Recognising that they are following an inflation target, and that they are currently on track, and so they will sterilize the bond purchase they are making from the government. The goal here is to avoid an increase in the money supply, and they will do so by increasing interest rates on the private sector.
2. They could abandon their inflation target, and make an unsterilized purchase of government bonds. In this context, this will lead to a short term increase in both inflation and output – but in the medium term output will ease back and inflation will remain higher and more volatile.
In the first case, the central bank sticks to its guns and will have to hike interest rates. The higher interest rates make labour, capital, and land available for the public sector to use when making its own goods and services.
In the second case, the central bank has broken their word about inflation targeting – however, everyone was choosing what they will do based on the expectation that the price level would rise at a predictable rate. Given this, the sudden lift in government demand places upward pressure on demand for labour, capital, land, and the commensurate private goods and services used to compensate government workers.
Given it is unanticipated, this is initially taken as higher demand for specific goods and services – leading to firms trying to get machines to run for longer and workers to stay in a bit longer than they would have previously. However, once it becomes clear that other prices are rising, firms pull back to more “normal” levels of activity.
It both of these contexts we have a form of tax, and both can be seen as a form of inflation tax – ultimately, the government’s decision that they would money finance works by “crowding out” both private sector activity and individual’s leisure/outside opportunities.
An important note about the higher level of output that occurs here – this is not necessarily a benefit. In fact, having output that is “too high” is a cost. Remember, people have to work longer hours based on the fact they expect higher remuneration than they end up getting. Firms are open longer expecting higher larger profits. Having to work longer is a cost, a cost that these individuals and firms were tricked into taking on due to unanticipated inflation. This work, for lower than anticipated real pay, is in fact part of the cost of the tax!
Note: Again, the central bank is managing the “demand” side of the coin here. Given it takes significant time for government programmes to get going, the fact that unemployment is elevated at a point in time does not imply there is a free lunch from money-financed expenditure now. However, this is an additional justification for having “automatic stabilisers” in place, such as a welfare system, as we do in New Zealand.
Loss of credibility, loss of benefit
The government above only temporarily increased spending, and the central bank only allowed a temporary one off increase in the money supply. As a result, why do we suddenly get higher and more volatile inflation?
Effectively, the central bank has lost credibility in such an environment – by showing it will bow to the whim of government financing, firms and households know there is some positive chance that the central bank will do it again. It will change expectations.
As a result, firms and workers will increase their prices and wages a bit more than the target rate of inflation by the central bank, in order to protect themselves against future central bank actions.
It is in this context that we get the social costs of inflation being more permanent than any perceived benefits. This is a compelling cost that exists in excess of other form of tax financing – and it is this credibility factor combined with the lack of transparency around a “money financed” form of taxation that leads economists to generally suggest other forms of taxation instead.
But it stops borrowing!
Wait a second here, we have to be clear about the difference between funding with borrowing and funding through taxation.
When the government borrows off its own citizens, it is increasing its claim on inputs, goods, and services now and reduce them again in the future. When the government temporarily increases taxes and then cuts them, it is increasing its claim on inputs, goods, and services now and then reducing them again in the future. In that way the difference between borrowing and tax is the determination of who is sacrificing inputs, goods and services such that the government can use/transfer them.
Some may say that it reduces borrowing from overseas – but this is a fallacious view, suffering from the “fallacy of composition”. In this way it is a much bigger logical leap than it sounds at first brush. Evidence suggests that government consumption and transfers, even when funded through taxation, lead to greater national borrowing.
Ramping up the inflation rate as a form of taxation
However, one off monetary financing is not the only way that the government can introduce a tax. Another type of inflation tax is called “seigniorage”.
In this case, a set inflation target could well be consistent and anticipated, but its existence cuts the value (in terms of real goods and services) of fiat currency at a constant rate. Greg Mankiw (1987) discusses the idea of optimal seigniorage, noting that it is a distortionary tax in the same way factor and consumption taxes are.
This implies that the choice of an inflation target is, at least partially, the choice of a tax rate in the New Zealand economy. As a result, the gradual increase in New Zealand’s inflation target has been akin to a slight increase in real tax rates.
[Note: There is the additional idea of fiscal drag which is separate from this. Fiscal drag captures how inflation and a progressive tax system interact. A progressive tax system has a number of “tax thresholds” that are set in terms of nominal income. As inflation increases nominal incomes (for the same real income in terms of the goods and services it can buy) this implies that more and more people will move into higher tax brackets – and so pay a higher income tax!]
Paying seignoiorage can be avoided by those who can protect their wealth and income. If the inflation rate can be guessed with a fair amount of accuracy, people will increase the price they charge for assets, and the interest rate they demand, given knowledge of this inflation rate. Workers will try to protect themselves by including some type of inflation adjustment in their wage demands. Most broadly this type of inflation tax works by pinging groups who are unable to easily pass on inflation when it comes to setting prices and wages!
A further important point here is that seigniorage works as a tax on “non-interest bearing assets”. However, since the 1980s the institutional structure of the banking system has changed a lot – it is now possible to get on call accounts that would offer a small rate of interest, a nominal rate of interest that could in turn adjust to account for rising inflation.
If we’ve move to a situation where all additional saving and borrowing is “inflation protected” then the ability to raise seigniorage revenue is becoming more limited.
It is always and everywhere about goods and services
No matter how much people talk about money, interest rates, inflation, and exchange rates these things only truly matter in terms of real goods and services.
The lens of taxation helps us to understand how different government policy influence the distribution of these goods and services, and the incentives people face to produce these goods and services.
Post-script: Given the subject nature of what I have written, I have to note something for those who are planning to say something such as “loans create deposits” and “the money supply is endogenous”. I appreciate where you are coming from, but I would note that this does not contradict the discussion above. Loans and deposits are created simultaneously due to supply and demand – and there is a dynamic process that occurs that defines where variables go following a “shock”.
The appropriate description we use depends upon the shock we are looking at. Above, we are specifically talking about government swapping borrowing for printing currency, given that capital, land, labour, technology, and entrepreneurship all exist for creating goods and services. In this context, and given our assumption that the central bank is following a clear and known rule, looking at the trade-offs inherent in the provision of goods and services through the lens of taxation is appropriate!