What is Inflation, and why is it Bad?
Most people think that inflation means a rise in prices, but that's not quite right. Inflation is not a one-off event, it's a process of continually rising prices that is usually measured at an annual percentage rate. It's an important distinction.
A Nobel Prize winning economist, and a firm favorite of mine, Milton Friedman is best known for his work on Monetarism and the role of the money supply in determining inflation.
He authored many books throughout an illustrious career, and you can find many hours of his brilliance on YouTube and elsewhere online. He was a prominent member of the Chicago School of Economics, a neoclassical school and adversary of Keynesian economics.
He was at his most influential during the Ronald Reagan presidency, but he was not a Conservative, he was a Libertarian who favored only a very limited role for government.
Technically, inflation was a word that was originally used specifically to describe the state of a growing money supply, and a rising price level of consumer goods was seen as a consequence of that growth. Some economists still conflate these two meanings, and it pays to be aware of that in order to avoid confusion.
On this website I use the word inflation in its modern sense, as relating directly to persistent rises in the general price level, regardless of the money supply, with consequent loss of purchasing power for any given amount of currency.
A price change to reflect market forces is usually desirable, but a higher inflation rate beyond some minimal target level is always a bad thing for society as a whole. It may be that at times a little more inflation is accepted as the lesser of two evils, but it is never a good thing in and of itself.
The cost of inflation to society comes in either its destabilizing effects on the economy, or in the unfair way in which it redistributes wealth. Both of these bad effects are explained in a section below, but first let's have a look at how the economic analysis of inflation has developed over the years.
Governments in the post World War 2 were focused primarily on reducing unemployment, and at the time it seemed that there was a tradeoff between either low unemployment with higher price inflation, or vice-versa.
This tradeoff was mapped out by the Phillips Curve, named after the economist William Phillips, and successive governments would simply choose where on the curve their tradeoff preferences were optimized, and then set their economic policies to reach that point.
The idea behind the tradeoff was that, when unemployment was low, firms would experience difficulty recruiting extra workers, and that consequently existing workers would have more power in the marketplace to negotiate higher wages. This would then create a kind of cost push inflation because firms would need to recoup the cost of those higher wages by charging higher prices for their goods and services.
Things seemed to be working well until the 1970s when, inexplicably at the time, the relationship between the unemployment rate and the inflation rate broke down. Persistent high inflation occurred at that time even as unemployment numbers increased.
The accepted wisdom of the day, that consumer price inflation was caused by cost push factors related to rising wages, had failed to explain what was happening. Economists went back to the drawing board.
It turns out that higher wage demands are not usually the cause of price inflation, they are more often a response to it.
A new theory of inflation was needed, and one was duly put forward by Milton Friedman that came to be accepted by the economics profession. In Friedman's own words:
"In order to stop inflation you have to have the government spend less and print less. That's the only way you are going to stop inflation, that's the one and only cure."
In other words, the money supply in an economy is the sole determinant of inflation. If it is increasing faster than economic output can keep up, the economy will overheat and inflation will result. Conversely, if it grows too slowly (as Friedman claimed was the case prior to the great depression of the 1930s) the economy will contract with growing unemployment and deflation (i.e. negative inflation) being the result.
These insights showed that there was no long-run tradeoff between unemployment and inflation, and that governments could seek to obtain an economy with both full employment and stable low inflation. However, to get the price level under control there is no way to avoid the pain that comes with a contraction in the money supply i.e. temporary unemployment during the adjustment phase.
In other words, in the short-run (which Friedman estimated to be about 2-years) the Phillips curve did exist and there was a tradeoff between price inflation and unemployment, but in the long-run this relationship disappeared, with unemployment in the economy settling at its 'natural rate'.
A key breakthrough in Friedman's model was the role of 'inflation expectations', and for full information about that have a look at:
Cost Push Inflation or Demand Pull Inflation?
The debate over how inflation arises and persists is not settled. Whilst the old theory of wage push inflation is somewhat debunked, it is quite clear that supply chain issues can cause higher costs of production for industry that lead to rising prices for consumers.
Demand pull inflation is however, the more typical type of inflation, and it fits nicely with the quantity theory of money that Friedman worked with to develop his ideas.
Increases in government spending, funded by money printing operations at the central bank, will lead to more consumer spending and higher gross domestic product. However, it is the sustainability problem that makes this a bad idea. Endless government spending just creates a long term debt crisis, and we can't spend our way out of a debt crisis. It just creates even higher inflation of prices in the end, with much more hardship to follow.
What is Inflation measured by?
The Consumer Price Index (CPI) is the internationally accepted measure of inflation, and it is this measure that is used by our governments and central banks as a target measure i.e. a low target level of around 2% per year is typical, and monetary policy is used to try and keep the actual CPI rate within close proximity of that target.
The consumer price index uses a representative basket of consumer goods which it amends each year to keep up to date with consumption patterns. As the prices of individual items change, the basket as a whole varies in cost, and the overall rate of change in that cost is used as the measure of inflation.
The basket clearly needs to be very large to be representative and reliable, and something like a thousand items are included in it. Each item is 'weighted' to account for relative importance e.g. pencil costs would be less important than wine costs because people tend to spend a lot more on wine each year than they do on pencils.
No price index can capture a perfect estimate of inflation, and no single figure can determine the correct policy action. The main problems with the CPI are discussed below.
Problems with the Consumer Price Index
Items in the CPI Basket
Following on from the 2008 financial crisis that hit the developed world like a sucker punch, questions need to be asked about why the financial authorities chose to ignore housing and real estate prices in the run up to the meltdown.
Of all the prices that change within an economic cycle, rising housing and real estate prices are arguably the most closely aligned with boom periods. Ignoring them (the CPI doesn't include them in the basket) seems naive at best, and negligent at worst.
Not only are these prices well aligned with boom periods, and therefore indicative of the need for a tighter monetary policy, it was specifically the inflation rate of these prices that led to all the junk bonds floating around the banking sector in 2008.
If something had been done sooner, the costly bailout of the banking sector could have been avoided.
Purchasing Power or Economic Performance?
If an estimate of purchasing power changes is all that is needed then the consumer price index may be an acceptable measure, but given that it is the measure that dictates monetary policy, I would argue that the primary purpose of the CPI should be to reflect the current state of a domestic economy's over-performance or under-performance.
For example, petrol/gasoline is included in the basket, even though its price heavily reflects international supply and demand factors for oil rather than domestic factors, meaning that it is best used as an indicator of how the global economy is performing rather than a domestic economy.
Granted, oil is a hugely important factor in the overall cost of production in an economy, and its influence needs to be taken into account, but it would be far better to do this separately rather than simply bundling it in with other items in the basket.
Conversely, housing costs are not included in the basket, even though they are an excellent item for judging the state of the domestic economy. Housing and real estate costs conveyed almost perfect information for predicting that our economies were heavily overheating in the run up to the 2008 financial crisis, but this information was discarded.
Many manufactured items are included in the basket, with no consideration for where those goods were produced i.e. in the domestic economy or in a foreign country. These items offer almost no credible value for estimating whether domestic economic growth is proceeding at a sustainable level or not.
Quality rather than Prices
Accounting merely for the price changes of items fails to appreciate changes in the nature of the items themselves. For example, it is a relatively straightforward affair to track how the price of entry level gaming laptops vary over time, but what about their performance? Certain items, and especially those that use/provide advancing technologies, evolve into very different items over time. Whilst the consumer price index does attempt to account for these changes, it fails to do so with any great degree of accuracy.
Even relatively old technologies can change significantly over time, and no one would argue that the cars of 20 or 30 years ago are comparable to the cars of today. Even basic family cars today are built with far superior handling, safety, performance, economy, reliability and so on compared to the cars of previous decades.
Clearly, changes in the quality of a product rather than the price of that product add an extra dimension of inaccuracy to the CPI. This is particularly problematic when using the index to estimate changes in purchasing power over time.
A final limitation of the CPI is the substitution bias that consumers have when prices change. Most goods and services have a range of viable substitute goods that could be purchased in the event that a price increase in any particular one of them makes them too expensive.
The substitution bias may have a temporary effect or a much longer lasting effect depending on the nature of the market (and price elasticity of demand), but in any case the CPI is not well equipped to track these sorts of consumer reactions, and to that extent the accuracy of the CPI is further compromised.