Perhaps no other issue is more hotly debated in the economics blogosphere than the discussion about inflation. Do we have too much or not enough? Will we have too much or not enough? What will the economy look like when these conditions exist? The stakes are high, not only for the real economy, but for bloggers and economists, who risk internet shaming and ridicule should they get their analysis wrong. Jokes aside, almost every major topic being discussed right now (sovereign debt, currency wars, etc) are at least indirectly related to inflation and its effects, so I think a comprehensive discussion of inflation/deflation is relevant.
The actual definition of the word ‘inflation’ has undergone multiple transformations over the last two centuries. Back in 1997, Mike Bryan, currently the senior economist at the Atlanta Federal Reserve Bank, put out a very good article on the history of the word, which I strongly recommend reading. According to the article, the word has basically gone through three phases in modern economic history. The first was in the 19th century with the classical economists, who largely used the word
…not in reference to something that happens to prices, but as something that happens to a paper currency.
“The astonishing proportion between the amount of paper circulation representing money, and the amount of specie actually in the Banks, during the past few years, has been a matter of serious concern … [This] inflation of the currency makes prices rise.”
—From the Bee (1855)
The second phase came about in the early part of the 20th century, when the meaning of the word had changed, to include the relationship between the money supply, trade needs, and the price level. The final stage, Bryan writes, came about with the Keynesian revolution, which completely separated the word from its monetary roots and redefined it solely in terms of prices. Bryan writes:
In addition to separating the price level from the money stock, the Keynesian revolution in economics appears to have separated the word inflation from a condition of money and redefined it as a description of prices. In this way, inflation became synonymous with any price increase. Indeed, Keynes spoke about different “types” of inflation, including income, profit, commodity, and capital inflation. Today, little distinction is made between a price increase and inflation, and we commonly hear reports of energy inflation, medical care inflation, and even wage inflation. Some go so far as to argue that the monetary definition forces the word to take on too specific a meaning:
“Even if we agree that an inflationary situation is to be taken to imply something about prices, precise definitions vary … Part of the difficulty here is that definitions of the more popular variety such as “too much money chasing too few goods,” not only purport to define inflation, but also imply something more about particular inflationary processes.”
—R. J. Ball (1964)
Two more modern definitions, first from Webster and then from Greg Mankiws popular college text Principles of Economics:
Inflation: a continuing rise in the general price level usually attributed to an increase in the volume of money and credit relative to available goods and services
Inflation is a sustained increase in the average level of prices.
So the definition of inflation has come to represent different things over time. The question, at least in my view, should be about whether or not the transformation over time has been a constructive one. That is, is the newer definition more helpful to us in analyzing the economy? Joe Wiesenthal of Business Insider, writing on Twitter, made these two comments:
FWIW: If inflation isn’t defined around rising prices/loss of purchasing power, then the term has no use to any real people.
— Joseph Weisenthal (@TheStalwart) February 21, 2013
@hblodget Right. How we always use it is the much more useful definition. But if I was wrong on the term’s history, I’ll accept that.
— Joseph Weisenthal (@TheStalwart) February 21, 2013
For a start, if the most common usage of a term is incorrect, or based on an incorrect concept, then it is not so useful. The definition of inflation as rising prices runs into two main problems. The first is that analyzing conditions from that definition involves a high degree of subjectivity. In the two most modern definitions above, the use of words such as ‘continuing’ and ‘sustained’ enable some wiggle room for the interpreter. How long is sustained, exactly? Three months, a year, five years? There is further subjectivity introduced in interpreting the actual rate of inflation itself. According to most commentators who use the inflation as price rises definition, it is not enough to have a positive rate of inflation. For it to be a good thing for an economy, the rate of inflation mustn’t be too low, or too high. Prices falling is out of the question (more on that later). On top of that, it shouldn’t fluctuate, but be relatively stable. Depending on who you ask, this magic threshold beyond which inflation becomes too much of a good thing differs. According to Federal Reserve Chair Ben Bernanke, a constant level of 2% inflation is what policy should aim for. (Yet another layer of subjectivity is introduced here because Bernanke prefers using the Personal Consumption Expenditures (PCE) rather than the Consumer Price Index (CPI) to determine his level of inflation.) Paul Krugman believes that at the moment a 2% inflation rate as too low, given the depressed state of the economy. A 3-5% inflation level wouldn’t be so bad in his view. Even higher levels of inflation pose no problems, according to Randall Wray, writing:
Economic studies actually find that the negative economic effects of moderate inflation are small—and by moderate I mean in the range of 10% to 20%; one famous study by orthodox economists even found only small effects up to an inflation rate of 40%. Indeed, the evidence is pretty strong that employment and growth do slightly better with moderate inflation than with extremely low inflation (although stagflation is also a possible result)
It’s conceivable then, that a given rise in the CPI would be treated differently by all three of the people I’ve just mentioned. The second main problem with the inflation as rising prices definition is the issue that interpreting data from that definition confuses cause and effect with respect to economic phenomena. Ignoring the subjectivity issue, simply noting that prices have risen and then making a determination about the health of the economy from that is similar to a doctor learning that a patient has a headache and from there advocating a particular course of action. Both the headache and the increase in prices are effects, the cause of which is what is actually important when making judgments going forward. Prices can rise for various reasons, including a decrease in the amount of goods available, or an increase in the money supply. In defining inflation as rising prices, no distinction is made as to why prices have risen. This is a problematic oversight because this ‘why’ is what really tells us whether the rise in prices is a good thing or not.
Consider a natural disaster that wipes out half of the goods and services in existence. While the reduction in supply of goods and services is undoubtedly devastating, the increase in prices of the remaining goods does serve two, positive purposes. It first discourages excess consumption of goods in a scenario in which goods are scarcer. Secondly it creates an increased incentive for producers to produce more goods. To reiterate, the scenario in which the supply of goods is halved is devastating and hopefully never occurs. The point is that once it has already occurred, the mechanism of rising prices has the positive effect of incentivizing increased production to replenish lost goods while also conserving the existing supply that wasn’t wiped out.
In the case of prices rising as a result of an increase in the supply of money, again consumption of goods becomes discouraged, as a rising price means that a higher percentage of ones’ income is needed to consume. The difference is that supply hasn’t been lost as before. Rather, assuming increased productivity, there are actually more goods available. So the discouragement to consume brought about by higher prices is hardly justified. The same incentive for producers to produce exists, which in the future will increase the supply of goods even further. Continually increasing the price of these goods only leads to a scenario in which goods and services begin to be priced beyond the reach of more and more consumers. In other words, the increased price denied the procurement of the extra goods and services produced by industry. This is otherwise said as a reduction in purchasing power, and undoubtedly a negative thing.
In defining inflation as price increases, this sort of ‘qualitative’ analysis of what is going on is almost completely ignored, and the negative effects of price increases as a result of money supply increases is excused. The reason, as touched on above when mentioning subjectivity, is that stable price increases are thought to be a positive for the economy. This dictum exposes two psychological barriers in mainstream economic which are both extraordinarily flawed but pretty much drive policy.
The first barrier is the idea that deflation, defined in the mainstream as falling prices, is always bad. Paul Krugman lays out this argument in this post. I’ll quickly counter his points below.
There are actually three different reasons to worry about deflation, two on the demand side and one on the supply side.
So first of all: when people expect falling prices, they become less willing to spend, and in particular less willing to borrow. After all, when prices are falling, just sitting on cash becomes an investment with a positive real yield
And when that happens, the economy may stay depressed because people expect deflation, and deflation may continue because the economy remains depressed. That’s the deflationary trap we keep worrying about.
The missing link in this argument is that it is goods and services – not money itself – that fulfill human needs and wants. Even if you expect the price of things to fall in the future, the fact that you may need the product in question right now means that you aren’t going to wait. For example, even if you knew that the cost of food was going to halve in two weeks, you aren’t going to starve yourself for those two weeks and risk serious harm to your body. In a less extreme example, if you have purchased a laptop or a cell phone or tablet recently, you’ve done so with the knowledge that in two or three months there will be a better product available for the same price or less. However, the fact that you need, or want that product today for school or work outweighs the fact that you could get a better deal by holding out. Said differently, this argument against deflation ignores the tradeoff between having a product now and having access to its benefits sooner, versus getting more for your money later. The needs and wants of individual consumers determine how much of one factor will be traded for the other. Krugman continues:
A second effect: even aside from expectations of future deflation, falling prices worsen the position of debtors, by increasing the real burden of their debts.
It is true that falling prices increase the real burden of debts. However, the real value of income is also increasing simultaneously such that the increased real value of debt can be better absorbed. In other words, even though past debts are increasing in value, the real increase in purchasing power means that future costs such as energy, food, entertainment, etc. can be purchased with a smaller amount of money. This enables the past debt to be dealt with. In the event that the amount of debt taken on can’t be supported when prices fall, the issue then is that the size of debt taken on was too great in the first place. Again, recall the fact that price changes are merely the effects that come about by changes in the real factors affecting supply and demand.
Finally, in a deflationary economy, wages as well as prices often have to fall – and it’s a fact of life that it’s very hard to cut nominal wages — there’s downward nominal wage rigidity. What this means is that in general economies don’t manage to have falling wages unless they also have mass unemployment, so that workers are desperate enough to accept those wage declines. See Estonia and Latvia, cases of.
Once again, the fact that real gains are made is left out of the discussion. The ‘downward nominal wage rigidity’ Krugman references actually bolsters this point, as it implies that other prices fall faster than wages do. This is a real gain in purchasing power for the wage earner. It is also not a given that wages have to fall at all. The fact that all other prices fall mean that input costs can still end up lower than the price of the final good – and given the product, the increased demand at lower prices can still result in high revenues for the company.
This last point makes the same appeal as Keynes did in advocating inflation to cure unemployment. The observation that workers wouldn’t accept the lower nominal wage needed to lower the unemployment level led to the solution of lowering the real wage through inflation. While this achieved the goal of lowering wages in real terms, the wage earner accepted it because in nominal terms the wage had not been reduced. Ultimately, this tact relies on the perception that the average worker is ignorant of the fact that he or she is losing purchasing power.
A more basic counter to the ‘deflation is always bad’ mantra in mainstream economics is to point out the discrepancy between that mantra, and the fact that human progress is inherently deflationary (using a price based definition of the word). The increased productive capacity engineered by humanity over time has led to an increase in the supply of goods and services. As I’ve mentioned throughout this piece, price changes are merely the effects of changes in real conditions. Human progress changes real conditions by increasing the supply of goods. Against the backdrop of a relatively stable money supply, prices should fall.
Adherence to the ‘deflation is always bad’ mantra would require monetary action to counter such a decline in prices. By using monetary policy to engineer rising, or flat prices where there would have been falling prices, a situation is created in which goods are made artificially scarce, given that some members of society may not be able to afford goods at the engineered higher price versus the price the goods would have cleared at sans monetary support. In short, society is prevented from transforming its own increased productivity into increased consumption for more people. With respect to the semantics argument from above, an inflation-as-prices-rising definition would see ‘no inflation’ if the price level had been engineered higher such that, for example, CPI ended up flat or slightly rising. However, due to the fact that a productivity driven fall in prices should have happened, the end result of the monetary policy is to leave prices higher than they would have been. This is a negative outcome in terms of the ability for society as a whole to consume goods, and also exposes the major shortcoming of the view that inflation should be defined as increasing prices.
The point that price changes are the effect of real changes in economic conditions cannot be overstated. There is a tendency for people to understand the isolated point that human progress drives prices lower, but to still have reservations with embracing it because there can be explicit casualties. For example, falling prices may expose producers who have incurred high production costs that cannot be recovered at the new, lower price point. This phenomenon may lead to layoffs and bankruptcies. My response to that reservation is that human progress has almost always involved sacrifice and pain. As time has gone on, entire industries have become redundant and unnecessary because newer, more efficient methods of doing things have been embraced. When the automobile was invented, would it have been wise for society to use its resources to prop up the dying horse and buggy industry so that the jobs in that field could be maintained? Of course not. The obvious improvement to society as a whole was unfortunate for those working in a redundant industry, but it is the price society must pay for advancement. This is the crux of the mental block regarding deflation.
Misunderstanding Economic Growth
The second psychological barrier is really an extension of the unwarranted deflationphobia described above. Part of the rationale behind fighting price declines tooth and nail is that it supposedly stifles economic growth. A stable rate of price increases promotes growth, as the story goes. This begs the question as to what economic growth is, and how it should be measured. The mainstream view usually correctly describes economic growth as the increase in the amount of goods and services available, but then erroneously relies heavily on the GDP metric to measure that economic growth.
The main problem with the GDP metric is that it is basically a measure of income. It measures the dollar figure of all of the transactions that take place. Therefore, if from one year to the next prices rise and those prices are supported by easy monetary policy, the GDP metric will rise, and economists will declare that the economy is heading in a positive direction. Lost in this analysis is the fact that a money supply induced rise in prices in the face of increased productivity is a negative development, as described above. Again, the cause and effects are conflated. Increased economic growth (increased goods and services) represents a real change in economic conditions, the effect of which should be lower prices. However, falling prices would have a negative effect on the GDP metric, given that the total nominal amount of dollars changing hands might decline. This mental block troubles economists when squaring the fact that while in actuality growth has increased (more goods and services are available to society), the chosen metric for growth (GDP) didn’t necessarily rise, and even fell. Simon Kuznets, one of the creators of the GDP metric, understood the pitfalls of overreliance on it to represent economic growth, stating:
… the welfare of a nation can scarcely be inferred from a measure of national income.
Indeed, the focus on national income has led economists to view obviously negative occurrences such as natural disasters, riots and oil spills in a positive light because of the fact that recovery from said events requires large outlays in spending, which in turn increase the GDP figure. The fact that these occurrences are objectively bad and society would be better off had they not happened is almost lost on them. When the first estimate of Q4 2012 GDP was negative, the fact that decreased military spending was a huge part of the decline caused outrage. The real implication, that the United States was spending less money building weapons and dropping bombs on people in distant lands, was apparently bad news, simply because the GDP metric had fallen. Similarly, it is a fact in mainstream economics that World War II propelled the United States out of the Great Depression, based on the fact that the GDP metric had been boosted. Again, the real implication, that the United States diverted massive resources to building tanks, bombs and fighter jets instead of home appliances and automobiles, is lost. In this way, this sort of overreliance on GDP is dangerous, as it leads policymakers to do any and everything in the name of GDP, often times to the detriment of the real economy.
Inflation, Debt Deflation and Policy
This brings us to the policy actions taken in the wake of the bursting of the housing bubble and the onset of the Great Financial Crisis. Across the Western world, a push to de-lever drove asset prices lower, which in turn put large institutions and households, which depended on rising asset prices, in a precarious position. The resulting responses to these realities included things like layoffs, bankruptcies and foreclosues.
The policy response to that development was expansionary monetary and fiscal policy. Within the mainstream frame of view, these actions were not only justified but necessary. Deflation loomed (which under this view is always bad), and GDP was falling (which also must never happen). Once again this rigid focus on metrics serves to neglect the underlying story beneath.
That story was that at the time of the bubble bursting, household debt had become too great to service, and the rising home prices subsequently could not be supported. This real change in the ability for households to accumulate more debt to support prices led to the subsequent drop in prices. Remember again that prices merely respond to real changes in economic conditions. Seeking to combat those price changes through expansionary policy changes the effects, but leaves the real causes of those effects intact.
In engineering prices higher in the face of a de-leveraging process which warrants low prices, policy ends up preserving the bubble conditions that existed prior. Instead of allowing home prices to fall to a market clearing level, they have been propped up, in turn propping up securities prices upon which the financial system still rests. The private sector, which is still overleveraged (as indicated by its ongoing attempt to de-lever further), must now re-lever and once again establish rising debt levels, so as to keep the prices of homes and asset prices rising. However, this re-leveraging slowly recreates the scenario which necessitated de-leveraging in the first place, namely an over indebtedness which never really was resolved in the first place.
I’ve found that in chatting with people, this line of reasoning does register as making sense. When discussing the real implications of what all this means, the mental blocks kick in and all of a sudden it becomes difficult to accept. It becomes inconceivable to ‘do the right thing’ and allow the debt deflation process to play itself out given that necessarily means bankruptcies, layoffs and foreclosures. Indeed, in a recent blog post Krugman blasts the ‘immorality’ of that viewpoint given the harm that would occur:
Think about that: he’s saying that ordinary workers and families who have nothing to do with financial speculation should suffer severely — because that’s what happens in a recession — in order to curb the irrational exuberance of a handful of incredibly well-paid financial industry types
Bear in mind that this is what everyone saying that we should tighten monetary policy now because of bubbles is really saying: that ordinary Americans should lose their jobs because otherwise the boys on Wall Street, bless their hearts, might get a bit overexcited.
The ugliness is awesome.
Even though he is focusing on the Wall Street irrational exuberance aspect, the basic idea is that preventing bubbles, or allowing them to deflate once burst, is immoral because ordinary folks may lose their jobs in the process. These concerns parallel the concerns brought up over the ‘downside’ of human progress in general. In the same manner in which the introduction of the automobile doomed the horse and buggy industry, the attempt to deflate bubbles so as to return to a sustainable level of debt and leverage also doom anyone dependent on continued unsustainable proliferation of debt. In both cases, a move from the less efficient to the more efficient state of affairs results in harm for some segment of the economy. In the specific case of the post 2008 attempted move to sustainable debt levels, the degree of harm is very high, owing to the greater portion of the economy which seems dependent on an unsustainable condition.
In either case, employing expansionary policy to prevent the immediate harm amounts to nothing more than squandering resources to forestall progress. Virtually any example of progress involves immediate short term pain or discomfort. Students bore themselves studying long hours to obtain knowledge. Athletes endure muscle destroying pain and fatigue in order to enable them to be rebuilt stronger and better. Cities require back breaking work and sacrifice on a large scale. Kicking drug addiction requires immediate physical and mental pain from withdrawal. Eradicating cancer can involve harsh and painful chemotherapy. Yet on the other side of all of those endeavors is a better state, progress. Partying instead of studying, not lifting weights and continuing the feel good highs of drug use are the Krugman-endorsed ‘easy money’ equivalents in the aforementioned examples. Is it not immoral to impose that on people, instead of allowing them to seek real progress? Is it not immoral to purposely divert resources to prevent a less efficient horse and buggy industry from collapsing, thereby stifling societal progress? If so, it is equally immoral to divert resources preventing unsustainable debt levels from being reduced.
Returning finally to the definitional issue from the beginning; defining inflation as a rise in prices and combining that with the view that prices must never fall enables the existence of potentially dangerous blind spots. Economic changes that warrant price declines are combatted as a result, and the elimination of price declines is seen to be a success of policy. Indeed, the ‘inflation is not a problem’ refrain is sung loud and clear by the likes of Krugman and Weisenthal, and is the prevailing view across the spectrum in mainstream economics. So why do we still have a sluggish economy at all? According to Irving Fisher, determining the origins of economic weakness was indeed an issue that economists grappled with when analyzing price changes. Writing in 1913:
If it can be shown, for instance, that today the good things of this world are becoming scarce on the one hand while money and its substitutes are not becoming plentiful it would be reasonable to conclude that the fault lies with goods and not with money. If, on the contrary, it can be shown that money and its substitutes are becoming plentiful and that goods are not becoming scarce, it is reasonable to conclude that the fault lies chiefly on the monetary side.
This observation is essentially the crux of this article. Seeing as productivity hasn’t declined, nor has the potential for most economies to produce declined, the problem seemingly is on the monetary side. Specifically, the increase in prices when economic conditions suggest drops in prices should occur. That phenomenon places an unnecessary restraint on consumption. Viewing ‘inflation’ as ‘rising prices’ misses this negative development, while focusing on the money supply, as the early economists did, makes it plain to see.