The cost of inflation
Note: Other posts in this discussion are available under the tag “inflation debate“.
Hi again everyone. Apologises for the delay getting this out, I have been sick. I’m still feeling quite sick, so if you read anything you think is wrong be sure to comment on it – as it probably is.
Without too many substantive criticisms of our view of what inflation is (yet) it seems that we can move on with our discussion on “re-thinking monetary policy“.
The next step in this process is to ask, what are the costs of inflation?
Previously we defined inflation as:
when the price of all goods consistently increase for no “real” reason
Given this definition, we have to ask are there costs associated with a consistent increase in the general price level?
Wikipedia discusses the effects of inflation here (*). Interestingly, the authour of the wikipedia article seems to be pro low levels of inflation, something I’m not as fond of – but this will be an issue we will discuss after explaining the costs.
Now, starting with the assumption that higher mean rates of inflation also imply higher levels of volatility in the inflation rate wikipedia provides us with the following costs (Note, I have only taken the costs that are relevant for our context):
- Additional uncertainty about future price levels,
- Unintentional redistribution,
- Shoe leather costs,
- Menu costs,
- Distortions in the relative price,
- The current rate of inflation helps to determine future rates of inflation (such that all the above costs will also be repeated in the future partially as a result of current inflationary pressures).
Now the “redistribution” card is the one most often used by inflation fighters – however, I think it is a slightly overplayed card. Although it is costly to fix superannuation and tax brackets too inflation, such a policy provides one way of limiting the redistributive impact government policy may have in the face of inflation. Furthermore, private agents will have the incentive to index to inflation as long as the transaction costs of doing so are lower than the benefit in that environment.
As a result, there is some “upper ceiling” on the costs from inflation from this category. The cost is sheer irritation, and possibly some confusion, rather than the perverse costs economists often paint it out to be.
The next most popular costs are 3 and 4 (shoe leather costs and menu costs). However, these have been empirically measured and are quite low. As a result, you must be wondering why my rhetoric is always so anti-inflation.
You’re problem with inflation is?
Costs 1, 5, and 6 are the costs of inflation that truly subvert the nature of the economy.
Cost one is uncertainty. If high inflation is variable (or as we will discuss, high inflation messes up the efficiency of the price signal!) then uncertainty creeps into both the firms and households decision making process. As people are implicitly risk averse, uncertainty has a direct cost on peoples welfare. Furthermore, uncertainty limits firms incentive to invest.
Cost five is the one that drives me the most crazy – and the one that most other people don’t seem to care about! This cost states that the relative prices of goods mess up in an inflationary environment.
Now I care so much about “relative prices” because for an economist, having appropriate “relative prices” is the thing we really want from the market. When relative prices are set up efficiently, we know that the allocation of resources will be efficient.
This is a cost you don’t even see – people can’t tell that the allocation of resources could be more efficient, however, this does not stop this cost of inflation being very real.
Cost six follows on from the fact that current inflation rates influence future inflation rates. As inflation is largely the result of agents price setting behavoiur, and as many prices in the economy are sticky, if people expect a significant amount of inflation in the future it can become a self-fulfilling prophecy.
This implies that elevated levels of inflation now will keep inflation elevated in the future – implying that we will suffer from the first five costs to some degree in the future as well.
Wikipedia says that there are benefits from inflation as well – maybe these outweigh the costs!
The Wikipedia article also states that some people benefit from inflation. For example, people that have borrowed money would benefit from unexpected inflation, as the real value of their debt would fall. However, if the we truly have “persistent” increases in the price level, the nominal interest rate and agents behaviour would change to take this into account – as a result, I think these benefits are over-played.
The other benefit that is mentioned stems from the relative price issue I mentioned before. Specifically, the article correctly points out that prices are more “sticky” downward, and as a result if the relative price of these sticky price goods is falling, inflation may help take us too a better set of relative prices.
However, this argument doesn’t entirely convince me. If some prices are “sticky” downwards, then we would also require the other prices to be stick upwards to avoid the relative price adjustment wouldn’t we? If one good is relatively more expensive then it should be, wouldn’t that lead to substitution towards other goods which would lead to the relative price adjustment?
Furthermore, who says that goods that are sticky downwards are falling in relative value? The prime example of a good that has a “sticky downward price” is labour. With skills and labour productivity increasing, won’t the relative price of labour rise over time (I guess this depends on whether labour or capital productivity is increasing more 😛 ).
Another benefit they raise is that inflation allows bank’s to have “negative real interest rates” as a way of stabilising economic activity. This seems silly, given that people have a positive rate of time preference there is a positive interest rate that will still “stimulate” economic activity. Making real interest rates negative seems like overkill in a stabilisation sense.
As a result, I’m not convinced that these “benefits” from inflation is completely sensible.
Conclusion
So, we have determined that inflation has costs now, and costs in the future. This will be important for when we move on to discussing the “trade-off” involved in targeting inflation. So did I miss/mis-represent any costs?
I mostly agree. I would add that another reason to not like inflation is that it is a non-transparent form of taxation, one that we need a strong framework in place to stop governments exploiting.
Also, I would say that while it is possible that borrowers and lenders could negotiate inflation-indexed terms, they almost never do, and most real world inflation episodes do result in a transfer to debtors (but I see this as a cost, not a benefit). Unless you want to argue that lenders build in a profit margin to compensate for the occasional inflationary episode.
I don’t follow your reasoning on the zero-bound for nominal wages, if real wages need to fall in response to a shock then inflation will get us there quicker, no?
“Given that people have a positive rate of time preference there is a positive interest rate that will still “stimulate” economic activity.” 0% interest rates in Japan never seemed very stimulatory. Taking the fact that real long-term rates in places like US have averaged around 2% (off the top of my head), you seem to be claiming that monetary policy never needs to be more than 2% below neutral, that seems like a bold claim.
Given your stance in this post, are you about to suggest that the the RB should have a 0% inflation target (or even a price level target)?
“Also, I would say that while it is possible that borrowers and lenders could negotiate inflation-indexed terms, they almost never do, and most real world inflation episodes do result in a transfer to debtors”
So you could justify that the transaction costs of indexation are higher than I implied in my post – yet another reason to dislike inflation.
“I don’t follow your reasoning on the zero-bound for nominal wages, if real wages need to fall in response to a shock then inflation will get us there quicker, no?”
I wasn’t disagreeing with the idea that inflation may help “quicken up” a relative price adjustment in the case of a specific shock (eg house prices) – I was disagreeing with the absolute nature of the statement that inflation helps fix relative prices. After all, if the relative price of the good that is sticky should be increasing, inflation will make the matter worse.
Furthermore, in the case of a price shock we may require both the good and its substitute to have sticky prices for the relative price adjustment not to occur – as the substitutes price could also change to achieve the right relative price! Think of the fixed price as a Numéraire, eg. wages become the standard of value you compare other goods to) then we will have a price level adjustment over all other goods – but this isn’t inflation .
“0% interest rates in Japan never seemed very stimulatory”
To be fair, Japan’s population is shrinking so they are a whole different kettle off fish. Given that the population is falling but the capital stock remains, there is going to be significant downward pressure on non-tradable inflation – thats just the way it is.
“you seem to be claiming that monetary policy never needs to be more than 2% below neutral, that seems like a bold claim”
As long as the economy is managed properly 2% below neutral seems fine to me 🙂
“Given your stance in this post, are you about to suggest that the the RB should have a 0% inflation target (or even a price level target)?”
A 0% inflation target in theory – however, the difficulty associated with differentiating relative price movements (given increasing quality) in the data implies that an inflation target of around 1% would be appropriate.
Never a price level target – I’m not saying me should react to shocks in the price level, only to growth in the price level.
We’re not really is a great deal of disagreement, I just think that the evidence that nominal wages are sticky downwards is very compelling, and that the theoretical story that these sticky wages lead to excessive business cycle fluctuations is believable. Hence I’m happy to have 2% inflation or so. Sure other prices could adjust to lower real wages, but you’re putting a lot of faith on the central bank to recognize that this is a price level adjustment and not a increase in inflation.
On Japan, don’t you think there is a problem where if deflation gets established, the central bank can lose the ability to lower real rates to a stimulatory level? For this reason it would seem we might want a buffer against the possibility of negative inflation rates becoming established.
I knew you were an inflation hawk, I didn’t realize how hardcore. I hope your explain for your readership that a 1% inflation target would be considered fairly unorthodox in contemporary monetary framework theory, and explain why you think the establishment has it wrong.
“I just think that the evidence that nominal wages are sticky downwards is very compelling, and that the theoretical story that these sticky wages lead to excessive business cycle fluctuations is believable”
Yes, I agree. However, I don’t agree that the relative price of labour will always be falling – if the relative price of labour is rising inflation actually moves us further away from the optimal set of relative prices, and so is a cost – as they are sticky upwards as well (although not to the same degree admitadly).
I guess I’m not convinced that inflation “oils the wheels” of industry as much as the Wikipedia authour does.
“Sure other prices could adjust to lower real wages, but you’re putting a lot of faith on the central bank to recognize that this is a price level adjustment and not a increase in inflation.”
I thought that a 1% target would give them sufficient room to move given the uncertainty about whether we are seeing a relative or general price shift – isn’t that what the RBNZ used for a long time (a 0-2% band, mid-point is 1%).
I may have over-sold my position here, I’m not trying to say we should go for anything far different from the current establishment in terms of targets. I was under the impression a 1% target was fairly standard – especially when used in conjunction with other core measures.
Furthermore, I haven’t actually mentioned the size of a target yet in posts, as I haven’t discussed trade-offs. I am happy to talk with you about targets and re-evaluate my own position.
Why don’t you write a post on contemporary monetary policy (or at least a blog comment) 🙂