The more inflation falls, the more it hurts.

Real theories of the business cycle predict that the public would feel worse off if inflation rose (due to supply shocks.)  In contrast, demand-side models predict that an unanticipated fall in inflation would make the public feel worse off.  That’s because demand shocks reduce inflation by shifting AD to the left, which also reduces real GDP, and hence real national income.  Even though inflation falls, NGDP growth falls even more rapidly.  With low demand for investment goods, real interest rates also tend to fall.

The following article suggests that the public regards this as a demand-side recession, where lower inflation actually seems like higher inflation:

WASHINGTON (AP) — Inflation spooked the nation in the early 1980s. It surged and kept rising until it topped 13 percent.

These days, inflation is much lower. Yet to many Americans, it feels worse now. And for a good reason: Their income has been even flatter than inflation.

Back in the ’80’s, the money people made typically more than made up for high inflation. In 1981, banks would pay nearly 16 percent on a six-month CD. And workers typically got pay raises to match their higher living costs.

No more.

Over the 12 months that ended in February, consumer prices increased just 2.1 percent. Yet wages for many people have risen even less — if they’re not actually frozen.

Social Security recipients have gone two straight years with no increase in benefits. Money market rates? You need a magnifying glass to find them. . .

The median U.S. inflation-adjusted household income — wages and investment income — fell to $49,777 in 2009, the most recent year for which figures are available, the Census Bureau says. That was 0.7 percent less than in 2008.

Incomes probably dipped last year to $49,650, estimates Lynn Reaser, chief economist at Point Loma Nazarene University in San Diego and a board member of the National Association for Business Economics. That would mark a 0.3 percent drop from 2009. And incomes are likely to fall again this year — to $49,300, she says.

Just to be clear, I am referring to the entire three year long recession and sluggish recovery, not the last couple months.  During recent months there has been a modest supply shock, due to events in Libya and then more recently in Japan.  It’s too soon to know the severity of these two shocks.

In the comment sections of posts there is a disturbing tendency of people to make the following mistake.  They notice that monetary stimulus often raises commodity prices more than other prices.  They notice that supply shocks often raise commodity prices more than other prices.  They infer that monetary stimulus can cause a supply shock.  In fact, monetary stimulus raises commodity prices if and only if it raises expected future output.  There is no other mechanism.  In contrast, supply shocks reduce expected future output.  Don’t confuse AS and AD shocks.

Update 3/20/11:  I obviously meant “real commodity prices” when I referred to commodity prices in the preceding paragraph.  Several commenters pointed out that one could find theoretical mechanisms by which QE could raise commodity prices and reduce AS.  In macroeconomics one can find theoretical arguments for almost any proposition.  In my reply to Statsguy I provided numerous reasons why I don’t think those contractionary mechanisms would be empirically important.


Tags:

 
 
 

20 Responses to “The more inflation falls, the more it hurts.”

  1. Gravatar of effem effem
    19. March 2011 at 10:02

    No other mechanism? Commodities are effectively a zero-duration asset – and there aren’t many of those. When the yield curve is very steep the value of low-duration assets will rise relative to longer-duration assets.

    Even if you dont believe the logic then just look at the facts. Gold has returned ~20% annualized when real rates are negative and ~2% annualized when real rates are positive.

  2. Gravatar of Andy Harless Andy Harless
    19. March 2011 at 10:08

    I don’t think it’s quite true that “there is no other mechanism.” If a monetary stimulus reduces real interest rates, then it increases the incentive to hold commodities in inventory (or to delay production/extraction). In theory this does mean that future output (or future inventory disinvestment) will be higher, but in the present, it looks very much like a supply shock: prices go up, and production goes down. There is a germ of truth in the view that blames QE2 for today’s rising commodity prices, which is what makes it so insidious. The 5-year TIPS is yielding a record low of -76 basis points. Personally I’d be happy to see the Fed push it even lower, as I think the benefit would outweigh the cost, but I’m not going to deny that there is a potential cost.

  3. Gravatar of Benjamin Cole Benjamin Cole
    19. March 2011 at 10:48

    Excellent commentary. I wish this was required reading, as they say.

    Also, as ever, there is politics afoot. You have the Glenn Becks of the radio-world honking about gold and inflation. BTW Beck’s show runs gold commercials–his sponsors have a stake in the inflation-scare. Amusing.

    The inflation-mongering fits nicely into a “Democrats cause inflation though profligate spending and easy money” stereotype.

    I am not a Democrat. I would cut entire “Democratic” departments, such as Education, HUD, and Labor. (BTW, I would also cut USDA, Commerce and slice defense spending by 75 percent).

    Sumner’s post on OSHA is very telling.

    But right now, there seem to be a Fox-network-type fever about inflation, and completely insane appeals by some that we need tight money.

    Of course, the Republicans would like to drive Obama out of office, so they can run deficits and funnel money to their clients, instead of the Democrats.

    We can just hope that somehow Bernanke stays clear of the fray. With Libya and Japan in the mix, we need even more QE, Now is not the time for the dithering, feeble monetary policy of the Bank of Japan.

  4. Gravatar of dirk dirk
    19. March 2011 at 12:26

    As long as home prices keep falling: http://www.calculatedriskblog.com/2011/03/corelogic-house-prices-declined-25-in.html

    it looks like a deflationary world to me.

  5. Gravatar of Mark A. Sadowski Mark A. Sadowski
    19. March 2011 at 12:51

    When I first saw this AP news story it really irked me. In my opinion it treated the situation like a great perplexing mystery, when it’s actually an easily understandable phenomenon.

    But it shows you that people’s perceptions can be skewed. I had a high school history teacher (Mildred “Mildew” Taylor) in the early 1980s who had actually lived through the Great Depression and yet was convinced that the entire period was cursed with high inflation. Similarly you find many people who are scarred by the experience of double digit inflation in the early 1980s to such an extent that they think it was the worst period in our nation’s history.

    Deflation underscores why the “Misery Index’ is so useless. In January 1932 yoy CPI was -10% at a time when unemployment was likely about 20%. Thus the misery index was only 10, lower than at any time in the mid 1970s to mid 1980s. Try telling that to the millions who went cold and hungry that winter.

    Similarly, my memory of the period of double digit inflation, when the Misery Index was well over 20, was that it was no big deal. My father had a 6% mortgage. Thanks to double digit inflation our family’s balance sheet grew very nicely in the early 1980s.

    Inflation is not the worst problem. Not having enough nominal income to pay your expenses is.

  6. Gravatar of effem effem
    19. March 2011 at 12:55

    “Similarly, my memory of the period of double digit inflation, when the Misery Index was well over 20, was that it was no big deal.”

    You may have felt very differently if inflation was all commodity prices and now wages/houses as we are seeing now.

  7. Gravatar of TravisA TravisA
    19. March 2011 at 13:28

    Scott, could you describe a bit more (including your assumptions) behind: “In fact, monetary stimulus raises commodity prices if and only if it raises expected future output.”

    Are you talking about the output of all goods or just commodity goods? And by ‘output’ do you mean nominal or real output?

    Obviously, in one model with completely flexible prices, monetary stimulus raises all prices but does not increase real output.

  8. Gravatar of Mark A. Sadowski Mark A. Sadowski
    19. March 2011 at 13:49

    effem,
    You wrote:
    “You may have felt very differently if inflation was all commodity prices and now wages/houses as we are seeing now.”

    To a certain extent that was my point. But even all the current hysteria over commodity prices is way overdone. According to the CPI food prices have risen roughly 4% since July 2008 or at an average rate of 1.6% annually. The price of wheat for example is a tiny fraction of the price of a loaf of bread in the US, as most of it is labor. Energy prices are off 12% from July 2008. Gas in my town is still 50 cents less a gallon than it was over two and a half years ago. Apparel is still about the same price as it was in July 2008. That’s because although cotton is actually a substantial part of the cost of a pair of jeans for example, the unit labor costs that have gone into a pair of jeans have plunged.

    The problem isn’t the soaring prices, it’s the sagging incomes, and that’s a result of depressed AD.

  9. Gravatar of Scott Sumner Scott Sumner
    19. March 2011 at 14:39

    effem and Andy, It’s partly a question of how we define easy money. I don’t view low interest rates as easy money, but rather a symptom of a weak economy. Throughout history, low real rates have often been associated with low NGDP. My view is that higher NGDP usually means both higher output and higher real interest rates. And I define an expansionary policy as one expected to boost NGDP. Could higher expected NGDP reduce real interest rates? Perhaps, but I think the effect would be trivial compared to the other mechanisms. But I suppose anything is possible in theory.

    In practice, monetary stimulus only explains a small portion of the recent commodity price upswing, which is mostly driven by conditions in developing countries. I am pretty confident that output is positively related to the amount of monetary stimulus (which would not be true if it was an adverse supply shock.)

    Benjamin, I agree that we need more QE, but it doesn’t seem likely that we’ll get it. And the second dose might well be much less effective.

    dirk, There are different results in different sectors. Overall inflation is low but positive.

    Mark, Good points.

    Travis, Yes, in a flexible price world all prices rise equally, and output doesn’t change. With sticky prices output rises, and commodity prices tend to be procyclical, even relative to other prices. So they rise more than consumer prices when output increases. Beckworth had a graph showing a very strong correlation between commodity prices and industrial production in developing countries. That’s what’s driving them.

  10. Gravatar of StatsGuy StatsGuy
    19. March 2011 at 18:22

    ssumner:

    “In fact, monetary stimulus raises commodity prices if and only if it raises expected future output. There is no other mechanism.”

    Why do you make such absolute statements needlessly?

    Andy Harless writes:

    “Personally I’d be happy to see the Fed push it even lower, as I think the benefit would outweigh the cost, but I’m not going to deny that there is a potential cost.”

    Nails it. (Of course I agree – monetary stimulus is better than the alternative, but at least let’s be honest about the cost)

    Scott, the absolutism of your arguments just opens you up to strawman critiques. The commodity carry trades are a central feature of finance in the past several years, and you aren’t even trying to listen to the arguments made by the other side. (not really trying, as in making a good faith effort)

    For example, you would argue that commodity price increases is a good thing, and the mechanism is an increase in expected future output (and hence demand for commodities).

    Let’s consider the other side, which has two arguments:

    1) To the degree markets expect future currency dilution, they shift to commodities as a store of value. In essence, this mirrors speculative demand in the IS-LM framework. As a currency substitute, the price of commodities can decouple (at least temporarily) from expected future supply. (Would you call this a bubble, or just rational markets?)

    2) Second, many market players believe long term supply of key resources (oil, and others and increasingly costly) is restricted, and the US EIA has recently admitted that previous reports which show positive expected liquid petroleum distillate production trends were wrong. If that’s the case, some resources are not easily replenished (long run future supply curve is steep), and aggressive monetary policy shifts future demand into the present without fully compensating by increasing future supply. Markets don’t want to consume today when they perceive future scarcity, and so they bid up the prices of commodities because they perceive future supply as DECREASING (at least, per capita for members of developed economies, as billions of people start competing for middle class consumption opportunities).

    You, however, are a technological optimist and scoff at the notion of a steep (or even contracting) long run supply curve. But what if you were seriously pessimistic about technology? What if you thought that the last 50 years have been a temporary blip bought with one-shot-resource-exhaustion, or that resistant bacteria were on the verge of winning the battle against antibiotics, or any other doomsday arguments? Often you’ve claimed that you pride yourself in really listening to other’s arguments and giving them the benefit of the doubt. Try it here.

  11. Gravatar of Jon Jon
    19. March 2011 at 18:40

    Scott writes:

    In fact, monetary stimulus raises commodity prices if and only if it raises expected future output.

    Scott, output usually refers to ‘real output’. Commodity prices will rise per the equation of exchange for any debasement of money unless that debasement causes real-output to fall. So, your claim here is misleading.

    Fact is that money debasement shows up in commodity prices because commodity exchanges are liquid and deep, and therefore prices are not sticky and are among the first to react to a change in the money supply, ceteris paribus.

  12. Gravatar of Scott Sumner Scott Sumner
    20. March 2011 at 05:41

    Statsguy, You said;

    “You, however, are a technological optimist and scoff at the notion of a steep (or even contracting) long run supply curve. But what if you were seriously pessimistic about technology? What if you thought that the last 50 years have been a temporary blip bought with one-shot-resource-exhaustion, or that resistant bacteria were on the verge of winning the battle against antibiotics, or any other doomsday arguments? Often you’ve claimed that you pride yourself in really listening to other’s arguments and giving them the benefit of the doubt. Try it here.”

    Let’s start with the fact that you haven’t listened to my argument. I’ve consistently argued the supply of oil is relatively steep, which is why increases in demand (from developing countries) has pushed prices sharply higher. I also wrote a post supporting Tyler Cowen’s Great Stagnation hypothesis. So I don’t see what any of this has to do with my argument here. I do think the SRAS is currently fairly flat, but only because we are temporarily below capacity. That’s an entirely different question form long run optimism or pessimism.

    I’ll grant you that I shouldn’t have said “no other mechanism” I should have said “no other important mechanism.” (and will be sure to do so in the future!) In economics, one can always create theories to justify almost any argument. For instance, one could argue that monetary stimulus raises expected inflation, higher expected inflation raises the real tax rate of capital, and a higher real tax rate on capital will reduce AS. So I can play that game too. The question is whether this effect is any more important than the one I just dreamed up. Here’s a number of reasons why I think the real interest rate channel is also probably unimportant:

    1. There is very little evidence that QE2 had a big effect on commodity prices. Beckworth had a graph showing that commodity prices closely tracked industrial production in developing countries. But real interest rates in the US are not strongly (and negatively) correlated with IP in developing countries (which would be required to make the argument that the IP/Commodity price correlation is spurious.)

    2. To the extent that QE2 did play a small role in the commodity price boom, it was probably mostly due to changes in expected growth in the US, (as we know that world growth is the dominant factor for for prices, it stands to reason that US growth would also be an important channel.)

    3. Another mechanism by which QE2 might have raised commodity prices is dollar depreciation. But since most imports are not commodities, and many commodities are produced in the US and exported, the net effect of dollar depreciation is strongly positive, even if we pay more for some imported commodities.

    4. The mechanism by which it is claimed that QE2 might reduce output is lower real interest rates, which allegedly encourages producers to leave commodities in the ground. Note that this argument works for oil but not so well for perishable food. And even for oil, the argument is very weak, and the sign is unclear. Recall that monetary stimulus can only reduce real rates through the liquidity effect, which is a short term effect. But we also know that spot and forward prices are linked through intertemporal arbitrage. Let’s say the liquidity effect lasts for three years, by which time all wages and prices have adjusted to QE2. In that case the real interest rate channel can have only a tiny impact on the spot price of oil–much too small to be of macroeconomic significance.

    5. Actual real rates probably fell less than measured real rates, as the inflation adjustment in TIPS occurs with a lag of several months.

    5. Not all of the actual fall in real rates was due to QE2, some seems clearly associated with the Libya and Japan shocks, factors having nothing to do with QE2.

    6. Monetary stimulus can also raise real rates in a ratex model.

    To summarize, the vast majority of the commodity price shock is due to IP in developing countries. That small part due to the US is mostly caused by factors like expected growth, and dollar depreciation, which are expansionary. In theory one can always find effects that cut both ways, for almost any economic question. But that doesn’t make them important. I’d be shocked if lower real rates in the US caused by QE2 pushed up oil by more than a couple bucks a barrel–way too small to be of macroeconomic significance.

    The other points you raise such as the the carry trade and/or speculative demand for commodities are either addressed by the preceding, or have no obvious bearing on my argument in this post.

    Jon, I should have said relative commodity prices. Nominal prices can rise without any change in output.

  13. Gravatar of anon anon
    20. March 2011 at 10:52

    OT, but Krugman has a new post up, discussing the Asian currency crisis as demonstrating the rationale for capital controls (IIRC, some New Keynesian/New Monetarist models agree on this) and calling for QE3: “[a] major surge in domestically-driven inflation “” in particular, a surge in wages “” would do it. And there has been an uptick in core inflation measures that is a bit of a surprise[.]”

  14. Gravatar of Morgan Warstler Morgan Warstler
    20. March 2011 at 14:07

    Interest rates cannot go up.

    If they do the debt service as % of tax revenue will immediately require cuts to entitlements.

    The path I see is this:

    1. Use deficits to end public employee unions.
    2. Use SCOTUS rule to end Obamacare.
    3. Use SCOTUS rule to allow unlimited corporate donations.
    4. Use new GOP president and congressional majority to cut entitlements in such a way as to effect the Likely Voters the least.
    5. Raise interest rates.

  15. Gravatar of StatsGuy StatsGuy
    20. March 2011 at 18:26

    Well, I do try to listen before attacking an argument. (Except when I read flawed econometric papers, since I really can’t help myself.)

    First, you should note that I mostly agree with you.

    Second, the arguments were not just related to QE2, but “monetary stimulus”, which includes stimulus from the US, China, EU, and others (massive russian devaluation, etc.). I am not so silly as to blame Bernanke for the world’s ills.

    some specifics:

    “the net effect of dollar depreciation is strongly positive”

    Er, I believe US commodity exports are smaller than US commodity imports if including oil. Weren’t oil and oil products something like 250-300 billion in 2009 and 2010? Also, don’t you need to account for pass through costs of imported manufactured goods, which should harm the balance since the US is a net service exporter?

    I’m not sure how one gets that the net effect of depreciation on short term commodity trade balance is positive. Should I be looking at different data??

    I’m not sure I understand the Beckworth argument fully, even though I agree with the conclusion (commodity prices in US are primarily rising due to competition with more rapidly growing countries). Need to think about the causal path, but am inclined to accept the evidence.

    On point 4) A three year liquidity effect is not inconsequential, since it’s quite long enough to affect core inflation, and we have repeated rounds of monetary stimulus… The doomsayers would argue that each round is increasingly short lived as commodity prices react more negatively due to accelerated depletion. (I don’t agree, but heh.)

    I agree with you on external shocks.

  16. Gravatar of marcus nunes marcus nunes
    20. March 2011 at 19:21

    StatsGuy
    At the time I expanded a bit on Beckworth´s post:
    http://thefaintofheart.wordpress.com/2011/02/06/bernanke-and-higher-food-commodity-prices/

  17. Gravatar of StatsGuy StatsGuy
    20. March 2011 at 21:07

    Thanks, Marcus. Nice timeline.

    I do have one question – if the “competiton from China” story is the answer, shouldn’t we see those currencies appreciating (even in relation to commodities)? In essence, that story says total world demand for commodities increases because of increased wealth (industrial production) in developing countries. However, it seems that commodities have increased in price in relation to developing currencies too. This is either a short term problem (i.e. temporary supply bottleneck until new production capacity installed), or speculation (possibly based on wrong but still powerful beliefs), or indicative of long term supply constraints.

    BTW, another signal that commodity supply constraints are non-trivial (though not necessarily the most important factor) was contango in certain commodities markets at times during the crisis.

  18. Gravatar of marcus nunes marcus nunes
    21. March 2011 at 04:19

    @Statsguy
    Countries like Australia, among others, have a currency that fluctuates according to commodity prices. If it fluctuates more or less may depend on the composition of their commodity exports and the changes in particular commodity prices. But that detail I don´t have.
    I don´t know about details such as contango.

  19. Gravatar of Scott Sumner Scott Sumner
    21. March 2011 at 07:52

    anon, The small uptick in inflation is not a surprise to those of us who worry about 99 week extended UI.

    I don’t think the Asian crisis shows capital controls work.

    Statsguy, If all countries do QE, and the world economy booms, then commodity prices are likely to go much higher. I agree. It is a given that with China and India moving so many people into the middle class lifestyle that world prosperity means higher commodity prices. I am willing to pay that price, and consider it a net positive. I don’t “worry” about net positives.

    Suppose 30% of imports and 3% of GDP are commodity imports. Then a 10% dollar depreciation costs us 0.3% of GDP in higher commodity prices. But the other 70% of manufactured imports also rise in price, which helps US industries competing against imports. And 100% of our export industries are helped, including many commodity exporters. It seems clear to me that in any reasonable model the totality of all those effects is likely to be positive for output (in the short run, naturally.)

    My point about the liquidity effect was different. I don’t doubt that core inflation can rise, but I was addressing the relative increase in commodity prices, due to lower real interest rates. That seems small.

  20. Gravatar of Scott Sumner Scott Sumner
    21. March 2011 at 07:54

    Morgan, That’s a very cynical policy.

Leave a Reply