Why does money matter?

Everyone but a few crazy post-Keynesians acknowledge it matters for NGDP. Everyone except a few crazy PKs and real business cycle types agrees it matters for RGDP in the short run.  The question is why?  Here are some theories:

1.  Tight money raises interest rates–which leads to less output.

2.  Tight money reduces the amount of the medium of exchange, which leads to less exchange and less output.

3.  Tight money raises the ratio of hourly wages to per capita NGDP, which reduces employment and output.

Nick Rowe thinks the second mechanism is especially important, whereas I focus on the third.  Most people like the first, but I’m not even going to bother with that one.   Here’s Nick:

The unemployed hairdresser wants her nails done. The unemployed manicurist wants a massage. The unemployed masseuse wants a haircut. If a 3-way barter deal were easy to arrange, they would do it, and would not be unemployed. There is a mutually advantageous exchange that is not happening. Keynesian unemployment assumes a short-run equilibrium with haircuts, massages, and manicures lying on the sidewalk going to waste. Why don’t they pick them up? It’s not that the unemployed don’t know where to buy what they want to buy.

If barter were easy, this couldn’t happen. All three would agree to the mutually-improving 3-way barter deal. Even sticky prices couldn’t stop this happening. If all three women have set their prices 10% too high, their relative prices are still exactly right for the barter deal. Each sells her overpriced services in exchange for the other’s overpriced services….

The unemployed hairdresser is more than willing to give up her labour in exchange for a manicure, at the set prices, but is not willing to give up her money in exchange for a manicure. Same for the other two unemployed women. That’s why they are unemployed. They won’t spend their money.

Alex Tabarrok has a very intriguing post that tries to find evidence for the “medium of exchange” theory of money’s importance.  He finds that barter increased sharply during the Great Depression:

Rowe’s explanation put me in mind of a test. Barter is a solution to Keynesian unemployment but not to “RBC unemployment” which, since it is based on real factors, would also occur in a barter economy. So does barter increase during recessions?

There was a huge increase in barter and exchange associations during the Great Depression with hundreds of spontaneously formed groups across the country such as California’s Unemployed Exchange Association (U.X.A.). These barter groups covered perhaps as many as a million workers at their peak.

That evidence seems to support Nick’s view.  Is there any way it could be consistent with my sticky-wage model?  Perhaps in the Great Depression many workers became so poor that they could not afford certain items unless they were first able to earn some “money” (actually wealth) by selling something they could produce.  Barter was a way of affording something that you otherwise could not afford.  (And maybe saving a bit on taxes.)

My theory is more ad hoc than Nick’s, but does have one implication.  In a modern recession you’d expect to see much less barter, as workers are not nearly as poor.  Most people in American have enough “money” (i.e. media of exchange) to buy something if they want it.  What usually holds them back in recessions is not the inability to find some media of exchange, but a lack of income or wealth.  Because we usually have enough media of exchange to make purchases, if wealth is much higher than the 1930s you’d expect barter to be much rarer.  And that is what Alex finds:

What about today? Unfortunately, the IRS doesn’t keep statistics on barter (although barterers are supposed to report the value of barter exchanges).  Google Trends shows an increase in searches for barter in 2008-2009 but the increase is small. Some reports say that barter is up but these are isolated, I don’t see the systematic increase we saw during the Great Depression. I find this somewhat surprising as the internet and barter algorithms have made barter easier.

So I think the evidence cuts both ways; and we are still left with two plausible hypotheses for why money matters.  (I’m calling the first one listed above implausible.)  It’s likely that both sticky wages/prices and a monetary economy are necessary conditions for money non-neutrality; with barter you can’t have the sort of wage-stickiness that would cause unemployment.  So it’s hard to tell which is the most important.

Perhaps I should say a few words about why I think sticky wages matter, and why other economists don’t agree.  The sticky-wage theory is usually assumed to imply that real wages should be countercyclical, at least during demand shocks.  I think they are somewhat countercyclical during demand shocks (Sumner and Silver, JPE, 1989), but the evidence is unclear because we lack good data for either hourly nominal wages or the price level.

The biggest drop in the monetary base in the past 100 years was in 1920-21.  That was also the biggest drop in the WPI in the past 100 years.  And that was also the biggest increase in hourly manufacturing wages divided by the WPI in the past 100 years.  Coincidence?  Maybe, but when the data is awful I look for shocks so enormous that even with awful data the underlying relationships should be visible.  For me that’s 1920-21, and also 1929-1939.  And both periods suggest that sticky wages are a problem.

The modern data is much less clear.  In my view the right way to define real wages is nominal wages over per capita NGDP.  Since nominal wages are sluggish, and NGDP is highly cyclical, that definition would make “real” (actually relative) wages look countercyclical.  Unfortunately, other economists would argue that this pattern doesn’t prove much, as it would be expected under almost any business cycle theory.  So we are back to square one.  I think sticky wages are the key to money non-neutrality, but can’t prove it to those who use more conventional models.

In the comments someone recently argued that I don’t rely on formal models.  Yes I do, but like Milton Friedman I like ad hoc partial equilibrium models, not the DSGE models that are so popular today.  One model for output and employment, another for monetary policy and NGDP, another for interest rates, another for exchange rates, etc.  Yes, you can try to put them all together in a general equilibrium model, but I don’t think macroeconomists know enough for those large abstract models to be useful.  The world is too complicated.  Better to stick with what we do know.

BTW, don’t confuse the issue of real wage stickiness and the natural rate of unemployment, with nominal wage stickiness and cyclical unemployment.  For instance this post Tyler Cowen discusses evidence that has a bearing on the real wage/structural unemployment issue, but no bearing on nominal stickiness and cyclical unemployment.  (That’s not to say real and nominal stickiness cannot interact to make the problem worse, they almost certainly do.)


Tags:

 
 
 

55 Responses to “Why does money matter?”

  1. Gravatar of TGGP TGGP
    23. March 2011 at 06:26

    I’ve linked this before, but no harm in doing so again. It contrasts the “Old Monetarist” preference for partial equilibrium to the “New Monetarist” (since Lucas) preference for general equilibrium, casting it as Marshall vs Walras. I can buy the Lucas critique and the idea that a model try to be consistent in how it treats the economy, but to a layman it seems like the issue of sticky wages is somewhat separable.

  2. Gravatar of Doc Merlin Doc Merlin
    23. March 2011 at 06:55

    From the description, in Nick’s theory, adverse money supply shocks, cause AS decreases, as each person’s supply curve shifts rightward. Thats /really/ heterodox.

    “One model for output and employment, another for monetary policy and NGDP, another for interest rates, another for exchange rates, etc. Yes, you can try to put them all together in a general equilibrium model, but I don’t think macroeconomists know enough for those large abstract models to be useful. The world is too complicated. Better to stick with what we do know.”

    I would argue those partial equilibrium models (most of them anyway) are also buck.

  3. Gravatar of Doc Merlin Doc Merlin
    23. March 2011 at 06:56

    Sorry, I meant LEFTWARD.

  4. Gravatar of jsalvatier jsalvatier
    23. March 2011 at 07:11

    All this time I thought you were talking about 2 in weird ways. Now I am a bit confused, could you elaborate on how you think NGDP is casual in addition to the effect of 2? What makes individual agents reduce their activities further? Do they look at NGDP and real wages and then reduce their market activities? You haven’t clearly distinguished your notions from the monetary disequilibrium process.

  5. Gravatar of Gregor Bush Gregor Bush
    23. March 2011 at 07:26

    “…I like ad hoc partial equilibrium models, not the DSGE models that are so popular today. One model for output and employment, another for monetary policy and NGDP, another for interest rates, another for exchange rates, etc.”

    I don’t know Scott. I’d say you think more in general equilibrium terms than you imply in this post. In partial equilibrium, fiscal stimulus will boost nominal GDP and therefore RGDP if wages and prices are sticky. But you often argue (correctly, I think) that the Fed has a particular forecast path in mind for output and inflation and that any shock to AD from fiscal policy will be offset by tighter expected future path of monetary policy. Furthermore, the market knows this, which is why the effect of fiscal policy on expected future AD, and therefore current AD, is very small (if not zero). Therefore fiscal policy can only be effective when the Fed decides to allow it to impact the future path of AD.

    This strikes me as the type of result you might get with a NK DSGE model with rational expectations and a central bank that is unconstrained by the zero lower bound (am I wrong about this?). I see more partial equilibrium thinking from those who argue that fiscal policy will be highly effective simply because the economy is in excess supply.

  6. Gravatar of Cameron Cameron
    23. March 2011 at 07:28

    I’m confused as well. How would #2 lead to a recession without price and wage stickiness? Wouldn’t a reduction in the medium of exchange simply lead to lower prices and wages?

  7. Gravatar of ssumner ssumner
    23. March 2011 at 08:03

    TGGP, Thanks for the link.

    Doc Merlin, Nick says a negative monetary shock reduces prices, but an adverse supply shock raises prices. He’s definitely talking about demand. But I do think the term ‘demand’ is misleading.

    Jsalvatier, When nominal wages rise relative to NGDP, firms cut back on hours worked. That’s what reduces output and real income.

    Gregor, Yes, I often think about indirect effects, but I typically don’t work with fully flushed out GE models, rather I use partial equilibrium models to evaluation those indirect effects in an ad hoc way.

    Cameron, Number 2 also requires wage and price stickiness, Nick is very clear about that point.

  8. Gravatar of Greg Ransom Greg Ransom
    23. March 2011 at 08:37

    Worth thinking about:

    The period of Kuhnian normal science is dead in macroeconomics:

    http://hayekcenter.org/?p=4501

  9. Gravatar of david david
    23. March 2011 at 08:50

    Don’t price stickiness and wage stickiness have substantially different behavior on “the ratio of hourly wages to per capita NGDP”? Nominal rigidities are not made equal.

    I am not sure that #2 and #3 are substantially different, actually, given that #2 is generally modeled to work via a price rigidity (or an expectation-of-prices rigidity) of some kind.

    On another note:

    (That’s not to say real and nominal stickiness cannot interact to make the problem worse, they almost certainly do.)

    My impression of the NK literature is that the definitive answer is that nominal (price) rigidities, combined with otherwise wholly perfectly neoclassical behavior, are not enough to account for observed variations in aggregate output and employment, given plausible levels of menu costs; thus there must be significant real rigidities out there. One amplifies the effects of the other, helpfully, so those real rigidities do not have to be massive to generate considerable barriers to price adjustment.

    Real rigidities fit uncomfortably in Rowe’s barter model, come to think of it.

  10. Gravatar of Benjamin Cole Benjamin Cole
    23. March 2011 at 08:53

    Another very enjoyable and insightful post by Scott Sumner.

    In my layman’s way, I often wonder about leaving bags on cash on sidewalks at night (in moderate-income neighborhoods, in amoral nations like the USA).

    The guy who picks up the cash-bag starts spending the money (in Japan, this does not work. The guy who picks it up returns it to the bank).

    Increased economic activity ensues (in the USA version), with the slight moral hazard that the guy who picked up the cash gets something without investing or working (kind of like every big-boobed blonde in the USA).

    Although we have had no increase in our productive capacity, out rises. Ths belies the premise, “We cannot print our way to recovery.” We can in certain situations. Now is one of those times.

    I keep coming back to this. If banks have lots of liquidity, but do not lend, then how do we spur the economy? Would it not be better to leave cash out at night, or run lotteries that pay back more, in cash, than they take in? Juice payouts at racetracks?

  11. Gravatar of Steve Steve
    23. March 2011 at 09:23

    This is slightly off-topic, but interesting:

    Allan Meltzer (Milton Friedman biographer, Carnegie Mellon professor, and WSJ Op-Ed writer) was just on Bloomberg. He asserted that QE2 had absolutely NO benefit in improving the economic outlook since last summer. He said the improvement occured because the Republicans won the election.

  12. Gravatar of Tom Grey Tom Grey
    23. March 2011 at 09:26

    Print the money — instead of taxing it! No need to leave it on the street. Have a payroll tax holiday of $1 trillion, and print that much to pay for it. Those workers will mostly spend it, tho many will also reduce their own debt/ increase savings; which should increase investment.

    #1 is not implausible — I challenge you to show any inflationary economy which reduced inflation without higher rates (=tighter money).

    In the current crisis, we have “tighter money” even with lower rates, because there was so much speculative money locked into MBS & CDS (on MBS) in the housing market asset bubble, which popped. How many trillions of “money wealth” disappeared in 2008-2009? And in many housing markets money wealth is still disappearing.

    Just as land is a specific form of capital for which economic generalizations don’t work the same, housing is a type of capital investment that, unlike money, is not really fungible. It’s too bad this reality means simplified models fail. But reality is like it is. Speculation/ investment in housing is different, and more different than the difference between investment in day stocks and investment in a new company.

  13. Gravatar of TravisA TravisA
    23. March 2011 at 09:28

    Scott, you focus on wage stickiness as the reason why there is excess unemployment. The mental model that I built up from reading you was that QE causes non-sticky prices to move first. Then some employers boost production and hire more people in order to take advantage of those suddenly higher selling prices and low wages. But in the comments to this post:
    http://www.themoneyillusion.com/?p=9277#comments

    you said, “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.)”

    So basically, you are saying that an increase in commodity prices are *not* a means of recovery. So I am totally confused. What *is* the step-by-step method by which QE affects NGDP on a micro scale?

  14. Gravatar of Luis H Arroyo Luis H Arroyo
    23. March 2011 at 09:59

    Very beutiful post, Scott. But I can´t see any contradiction between N.Rowe and you. Tight money has both effect, I suppose. I don´t know what is the first having contractionary effects.
    In any case, your (partial) model is very clear. The problem is that in Europe we “confuse the issue of real wage stickiness and the natural rate of unemployment, with nominal wage stickiness and cyclical unemployment.”
    So, all the EU measures to “save” the Euro are namely tergeting the problem of real wages=structural unemployment. No word about the cyclical problem. I don´t see a debate about the different models, but a complete confusion. However, if cyclical problems is not fixed, there would be always frozen from a cyle to next.
    Nice post.

  15. Gravatar of Morgan Warstler Morgan Warstler
    23. March 2011 at 10:07

    In the modern age, Uncle Milty would prefer an Ebay/Paypal auction model attached to a Guaranteed Income. Inside of a month, we’d have 20M working… a lot more efficiently than any barter site.

    The other big issue for me is capital gains. We need to recognize that SMB owners are investing in themselves, and should be free to take profits from one of their plays and move them tax free into another newco.

    Right now SMB owners have an large incentive to keep growing their current business even if it has plateaued – taking out profits is income.

    Newcos account for all the new job growth, and they are all spawned out of the SMB crowd.

  16. Gravatar of Benjamin Cole Benjamin Cole
    23. March 2011 at 10:08

    Steve-

    Lately, I have been following Allan Meltzer and John Taylor, both very smart guys and recognized economists. I even exchanged some e-mails with John Taylor, and he is a gentleman.

    Nevertheless, I think both are rabidly partisan (in gentlemanly ways). They want Obama out, by any means necessary.

    This partisanship tends to color their position on anything as long as Obama is in office, and will color their position on anything as soon as the R-Party regains control.

  17. Gravatar of Benjamin Cole Benjamin Cole
    23. March 2011 at 10:10

    Steve-

    Just as when you hear R-Partiers talk about “impediments” in the economy, they never mention ethanol, now in your gas tank everywhere in the USA, courtesy of extraordinary subsidies and regulations. Oh, that.

  18. Gravatar of Benjamin Cole Benjamin Cole
    23. March 2011 at 10:11

    Tom Grey-

    Excellent suggestion. I like a payroll tax holiday.

  19. Gravatar of Morgan Warstler Morgan Warstler
    23. March 2011 at 10:17

    I don’t think we can call barter increase a proof of money demand.

    I think we can call it people have the need for DISCOUNTS.

    Barter is no different than Groupon, where you both buy and sell – have both excess inventory and are willing to discount and have things you will buy if they are cheaper.

    This gets to my main complaint with liberals: If you are arguing for ANY policy at all that promotes sticky wages, you cannot argue sticky wages as a reason for printing money.

  20. Gravatar of Morgan Warstler Morgan Warstler
    23. March 2011 at 10:19

    Barter = Discounts

    To be clear, this is because there is a subset of total things that people have excess of…

    There is no barter demand for Web2.0 coders right now. They have no excess, they are making more $ every 3 months.

    Barter is just available within a class of things that are mis-priced.

  21. Gravatar of Philo Philo
    23. March 2011 at 11:00

    I found this post puzzling. Explanation #2. (“Tight money reduces the amount of the medium of exchange, which leads to less exchange and less output”) seems lame on its face: tight money reduces the amount of money *relative to demand*, but not necessarily absolutely. And I do not see why you attribute #2. to Nick Rowe, rather than #3. (“Tight money raises the ratio of hourly wages to per capita NGDP, which reduces employment and output”). Rowe’s story is surely a sticky-something story: the hairdresser, manicurist, and masseuse are all charging too much for their services–prices/wages that were appropriate before but are too high now, given deflation or tight money. (Since they are self-employed, it seems rather arbitrary whether you call the sticky amounts they are trying to charge “prices” or “wages.”) So his story fits pretty well under 3. And why do you think the increase in barter during the Great Depression supports 2. *as against 3.*?

    You say: “What usually holds them back in recessions is not the inability to find some media of exchange, but a lack of income or wealth.” But they do have (potential) labor and goods; however, they refuse to exchange these for whatever money they could get–their prices/wages are sticky. They *could* find (= acquire) *some* media of exchange, but they refrain from doing so–they’re sticky.

    Aside: Sticky prices/wages must be important apart from monetary considerations. Values are continually changing; people’s supply and demand schedules are always in flux, at least somewhat. Slow-adjusting prices will prevent equilibrium from being achieved, perhaps even from being approached, even if the monetary system is functioning perfectly. Perhaps you have this in mind when you write: “That’s not to say real and nominal stickiness cannot interact to make the problem worse, they almost certainly do.” But I don’t see two different kinds of stickiness, “nominal” and “real”; and what sort of “interaction” do you have in mind?

    Altogether a puzzling post.

  22. Gravatar of StatsGuy StatsGuy
    23. March 2011 at 11:16

    I’m a little unclear how the three theories separately impact asset prices, and in particular relative solvency. Let me advance a silly notion (a fourth explanation) – tight money does not merely reflect increased risk premia, but actually creates increased risk premia. Yah, weird, huh?

    But let me observe something really strange. It wasn’t until AFTER tight money crashed the stock market that perception of the riskiness of stocks (and other asset prices) increased. Did _actual_ risk increase? Actually, _actual_ risk probably decreased after prices declined, but perceived risk is higher.

    This doesn’t fit into any models out there because it posits that preferences respond to money (rates or prices); the whole thing becomes non-stationary.

  23. Gravatar of david david
    23. March 2011 at 11:19

    This gets to my main complaint with liberals: If you are arguing for ANY policy at all that promotes sticky wages, you cannot argue sticky wages as a reason for printing money.

    Yes, you can. 😛 Having more sticky wages can be the second-best outcome, with completely flexible prices as the first-best outcome.

  24. Gravatar of anon anon
    23. March 2011 at 11:29

    Morgan Warstler, Groupon is not about barter in any sense. In fact, it’s not even about discounts in the conventional sense. The innovation in Groupon is that people have to sign up for coupons in advance, and the coupons are only issued (and paid for) if enough people also sign up. This is great for retailers since it enables them to have a guaranteed Black Friday whenever they want (the Thanksgiving kind, not the 1929 kind), so Groupon.com charges very high margins for this service.

    Another application of this same principle is provision of creative goods, as seen e.g. on Kickstarter.com. It works in the same way: people sign up and pledge to pay for the good, but they aren’t actually charged unless enough people sign up for the same pledge. By using such systems, many public goods can be provided on an entirely volunteer basis, with no need for government involvement or intrusive policies such as intellectual property restrictions.

  25. Gravatar of dlr dlr
    23. March 2011 at 11:50

    I feel like you might be missing a useful distinction (or maybe I’m misunderstanding). I read (2) to reflect a view that attributes recessions solely to a loss in production from less trade, due to insufficient lubrication (MOE). So it would look like the kind of drop in real GDP you might see in a Barter system if a law was passed to require people to jog 5 miles before engaging in any act of trade. It wouldn’t necessarily cause unemployment or underutilized capacity, but it could cause a technical recession (production decline).

    If that’s right, I think it can be part of the story of some New Keynesians, but isn’t what Nick is saying. Nick is always focused on the special aspect of demand-recessions being the fact that everybody wants to sell while nobody wants to buy. So he wouldn’t just say that “tight money reduces the medium of exchange, which leads to a lack of exchange,” but rather “tight money leads to an excess demand for money, which is ALREADY a recession.” That is, the very act of people demanding more money than there is (only possible with sticky prices) equals freezing in a monetary economy, and when people freeze then resources are idle. It is not about the inadequate buying AND selling of money that you seem (to me) to suggest, but only the inadequate selling of money.

    Now, I don’t really know what Nick thinks about what you might call stage two of recessions, when demand is not plunging but under-capacity emerges for a period of time. This excess demand for money leads to frustrated deflation under sticky prices and so all prices are too high relative to money, but surely some more than others, like wages. So after the initial plunge Nick might be on board with (3) as an additional reason, but might equally describe the same effect (high real wages relative to other prices and relative to real money balances) as merely a quasi-equilibrium reached amid a continuing excess demand for money. You can imagine the same situation described either way.

    Meanwhile, I think there are many NK who would dispute the assumption in your title that everybody thinks money “matters,” taking a cue from Woodford’s cashless economy. They would argue that it is not money that is driving anything at all, but that all three of your reasons play a role in recessions, but can be restated in terms that eliminate the importance of money, depending on the actual recession. They might also restate Nick’s reason as one in which money doesn’t really “matter” which is the key disagreement between the Rowians and the Woodfordians.

  26. Gravatar of very good post « Economics from Spain very good post « Economics from Spain
    23. March 2011 at 12:04

    […] Read http://www.themoneyillusion.com/?p=9335 […]

  27. Gravatar of jsalvatier jsalvatier
    23. March 2011 at 12:15

    @ssumner: that’s a description, not an explanation. Are you saying that they cut back on hours in addition to any cutbacks they might do because they lack money (mechanism 2)? What causes them to do this? This is what is not clear.

  28. Gravatar of Morgan Warstler Morgan Warstler
    23. March 2011 at 12:45

    anon, you miss my point – much like GILT Group, Groupon functions to fill the gaps (at a discount) for the retailer – the “deal tipping” (amount of people required) is a throw away.

    This is what barter really means in modern terms – you are a hair stylist, you have excess time to cut hair – you WISH you could do sell it for your retail price – you cannot, so you are willing to barter.

    We are not talking about caveman barter where there is no money, we are talking about barter in a money system where people have excess inventory.

    This is:

    a) a limited subset of the economy – only some people have inventory to sell/trade at a discount.

    b) NOT a good argument for printing more money. It is a good argument for busting up prices as much as possible.

  29. Gravatar of david david
    23. March 2011 at 12:52

    It is a good argument for busting up prices as much as possible.

    Of course, if you wish away price rigidity, then the problem goes away! Who knew.

  30. Gravatar of Morgan Warstler Morgan Warstler
    23. March 2011 at 13:12

    I don’t wish away price rigidity. I refuse to let DeKrugman argue both for public employee unions, Davis-Bacon, Minimum Wage, etc. and for printing money. You should too.

  31. Gravatar of ssumner ssumner
    23. March 2011 at 13:24

    Greg, I hope this is the end of normal macro, as this crisis provides a wake-up call that we can’t rely on a “wait and see,” centrally-planned stabilization policy. Time for letting the markets run the macroeconomy.

    David, Yes, both 2 and 3 (and indeed 1) definitely rely on wage/price rigidity.

    The question of “plausible menus costs” is interesting; one problem may be that we don’t know why wages and prices are sticky. We assume it’s menu costs, but it might be money illusion, or something else. And we don’t really know what sort of menu costs are plausible. They don’t need to be all that large.

    Benjamin, Or just stop paying interest on those bank reserves.

    Steve, He’s wrong, rumors of QE2 clearly drove equity prices higher in Sept./Oct. 2010.

    Tom, No need for deficits, just do monetary expansion.

    Travis, What I meant is that the QE2 was way too small to account for the huge run-up in commodity prices in recent months. I agree that QE2 did raise commodity prices modestly in September/October, but not since. Instead, other factors have raised them in recent months. Thus QE2 might ahve raised them say 10%, out of a total 50%. increase. That’s what I meant.

    Luis, Thanks, I don’t know if there is a contradiction, it’s more a question of which effect is most important. Perhaps both must be true.

    Morgan, We need to abolish capital gains taxes for everyone, and I agree that we need to oppose any policy that makes wages stickier.

    Philo, Nick and I have debated this beofre, and he certainly agrees that our two mechanisms are different, and he likes #2 better than #3.

    Nominal wage stickiness occurs when an hourly wage rate is negotiated at the equilibrium value, but then is kept fixed at that nominal value for a year. Real wage rigidity occurs when a union negotiates a wage 20% above equilibrium, and then adjusts it upward each time the price level rises.

    Both occur if a union negotiates a real wage 20% above equilibrium, and then keeps it fixed in nominal terms for a year.

    The same is true of minimum wages, etc.

    Statsguy, When you say risk premia rise, are you simply saying asset prices get more volatile, or are you saying that a given volatility of asset prices now requires a higher rate of return, as investors are now more fearful or risk?

    David, That’s what Krugman suggests, but the evidence from the 1930s contradicts him.

    dlr, That’s a good point, I think I did leave the wrong impression of Nick’s argument. I’ll have to think about this issue a bit more, and see if I can better describe the difference. Sometimes I think the real difference is not MOE vs. sticky wages, but sticky prices vs. sticky wages.

    Regarding Woodford, a long time ago I read one of his “moneyless” papers, and saw an economy that did in fact have a medium of account, which I would call “money.” My intuition might have been off, but Bennett McCallum had the same reaction, and he has the best intuition of any economist that I have read.

    So I still don’t know what a moneyless economy is. If it’s an economy without a medium of exchange, or where the price of the MOE differs from the medium of account, then I agree that it is possible. In that case I think the medium of account is the key, and the medium of exchange doesn’t matter. Thus if the Canadian dollar was used as the MOA in Canada, and all Canadian wages were nominally sticky in Can$ terms, but almost all transactions in Canada were done with US dollars, then the Can$ would be the asset of interest to monetary theorists. It would be tight Can$ policy, not tight US$ policy, that would cause recessions.

    jsalvatier, Businesses would cut back on hours if the wage/NGDP ratio increased due to tight money. I am assuming that in the short run hours worked are determined on the demand side of the labor market. Workers work as much as their bosses ask them to work. That drives changes in production, which drives changes in sales. Of course in a ratex model the cause can occur after the effect.

  32. Gravatar of Greg Ransom Greg Ransom
    23. March 2011 at 13:57

    It looks like Blanchard & Stiglitz and other powers that be want to go in another direction.

    Scott wrote,

    “Greg, I hope this is the end of normal macro, as this crisis provides a wake-up call that we can’t rely on a “wait and see,” centrally-planned stabilization policy. Time for letting the markets run the macroeconomy.”

  33. Gravatar of anon anon
    23. March 2011 at 13:58

    Morgan, that’s a discount. Calling it barter is very confusing, especially in the context of Rowe’s argument which is about non-monetary exchange. It’s clear that discounts are a partial solution to recession, because they increase effective demand in real terms through their effect on the price level (and aggregate supply). But since nominal rigidities are pervasive in the economy, monetary expansion is also needed.

    I agree about the way retailers are currently using Groupon, and how the “deal tipping” feature has seemingly become a gimmick. But this will probably change in future, especially as the economy improves and the site faces more competition.

  34. Gravatar of jsalvatier jsalvatier
    23. March 2011 at 14:06

    @ssumner,

    Your story seems identical to Nick Rowe’s monetary disequilibrium story only less clear. Unless you’re suggesting that businesses literally look at NGDP numbers (which aren’t even available in real time) as opposed to the demand for their products, sales etc. . NGDP *per se* can’t affect businesses decisions given that they cannot observe NGDP.

  35. Gravatar of Nick Rowe Nick Rowe
    23. March 2011 at 15:20

    Thanks Scott!

    Yep, 2 is definitely me.

    Here’s a thought experiment: Start in equilibrium. Hold all prices and wages fixed. Reduce the money supply. You get my story (2). Now let prices fall, but hold wages fixed. You get Scott’s story (3). But nothing really changes as you move from 2 to 3.

    dlr: Is Woodford’s “cashless economy” really a barter economy? Something I have never managed to get straight. Because if it is a barter economy, then he’s got my 3 women leaving haircuts, manicures, and massages on the table without picking them up — unexploited gains from trade in his equilibrium. Which is just wrong. Forget the medium of account role of money in his model. Think about the medium of exchange. Is monetary exchange “essential in his model or not???

    If it is essential, it’s an implicit assumption. If it’s not essential, he’s wrong (IMHO).

  36. Gravatar of Scott Sumner Scott Sumner
    23. March 2011 at 16:49

    Greg, Unfortunately.

    Jsalvatier, Businesses do observe NGDP even before the government does. But they observe it collectively, not individually. In other words, each business observes the demand for its product, and they know their wage cost of producing. If their wage cost rises relative to the demand for their product, they lay off workers. Collectively all these individual decisions add up to business (in the aggregate) responding to NGDP, and laying workers off if wage rates rise relative to NGDP.

    Nick, You said;

    “Here’s a thought experiment: Start in equilibrium. Hold all prices and wages fixed. Reduce the money supply. You get my story (2).”

    Yes, but you get mine as well, as I define relative wages as W/NGDP. So if monetary contraction reduces nominal spending, and prices are fixed, you still get a rise in relative wages (from lower Q.) As I said, I don’t think any other economists will buy into my definition of relative wages, but if NGDP is one’s nominal aggregate, it seems the only sensible way to look at the situation.

    I am pretty sure Woodford’s “moneyless” economies have money, he just calls it reserves. That’s the medium of account in terms of which he assumes price stickiness.

  37. Gravatar of Morgan Warstler Morgan Warstler
    23. March 2011 at 18:06

    anon, as I said, Alex and Nick both approached the barter idea as if it shed light on whether we need to print money.

    It doesn’t.

    It shed light on whether we need to have people sell off their excess inventory at a discount AND get to buy up other people’s excess inventory at a discount.

    It is not confusing.

    WHAT IS CONFUSING, is pretending that since many low wage types are all idle (our top 4 deciles are not suffering unemployment)… that since they would all like to barter with each other – we should PRINT MONEY.

    That’s headassbreathe confusing.

    —–

    So I stepped in to provide some clear modern explanation of our we see excess inventory being liquidated using the Internet as clearinghouse – in much the same way that the bartering they are discussing. You are welcome.

    —–

    Last note: I don’t think it is ever going away. The whole “tipped” deal is important – what is important is ensuring you secure enough foot traffic (altho I think the keep rate for Groupon has to go down), or like GILT, you are sure you can move most or all of your final inventory at production cost…

    This essentially lets you run a “club rate” and a retail or “impulse” rate. Having half your shop full of discounts only becomes a problem if they they crowd out your retail side.

    I’ll go ahead and pimp Guaranteed Income again:

    http://biggovernment.com/mwarstler/2011/01/04/guaranteed-income-the-christian-solution-to-our-economy/

  38. Gravatar of dlr dlr
    23. March 2011 at 18:22

    Nick/Scott,

    In the cashless limit, there is no medium of exchange to speak of. I think of the Woodford cashless economy as presuming that only part of the double coincidence problem is solved. Namely, the problem of traders finding and trusting each other. But not the computation problem. The payment technology can match buyers and sellers by item and even quantity, but still requires a medium of account to make markets – it cannot determine how many haircuts would be traded for one hour of manufacturing labor unless it has medium of account prices to reference from the points of exchange. To me, that’s how you get a unit of account and medium of account with no medium of exchange. Is that monetary? Unitary?

    I think you would say that resorting to pure Barter is still not possible here. By pure Barter, I mean, that which needs neither a medium of exchange nor medium of account. Let’s say that a worker and barber and factory are all willing to trade 20 minutes of labor (to the factory) for a haircut (to the worker) and a widget (to the barber). And the price of each is 20. Now shock the price level, such that the haircut and widget cost 10 but workers still cost 20 (but individual workers would still trade the 20 minutes of labor for a haircut). The payment system can’t overcome this, and the payment system is their only method of semi-barter. So they are just as stuck as they would be with a MOE.

    So in Wooodford’s unitary economy, Scott’s kind of recession (real wages are too high relative to other prices or even expected “NGDP”) can happen and Nick’s cannot – there is no medium of exchange that anyone has any demand to hold, so there is no way to freeze that replicates an excess demand for money. There would be all sorts of problems, I imagine Nick saying, but they would look like a bunch of excess demands and a bunch of excess supplies akin to what would happen in a pure Barter economy with screwy prices, as opposed to one big avalanche of sellers that you see in 1929 or 2008. I think Scott might call this economy monetary while Nick would not.

    The Central Bank would conduct “monetary” policy, or “unitary” policy, or maybe “intertemporal pricing” policy, using interest rates on central bank liabilities (the medium of account) as the instrument. The liabilities would be held only as securities (no MOE function) and would pay interest. You can see how this further blurs the continuum between monetary and fiscal policy.

    But in some meaningful way Woodford might say money no longer matters in this world. At least, the demand for the medium of exchange and the excess cash balance mechanisms do not exist. Nor does the expected future supply of the medium of exchange. Of course, that is also a way of saying they might not necessarily be important in recessions in a monetary economy, either.

  39. Gravatar of Mark A. Sadowski Mark A. Sadowski
    23. March 2011 at 19:26

    Scott,
    I don’t dispute your story on wages but when I think about “laboratory experiments” on tight money they usually involve explanation #2. My favorite is “The Economic Organisation of a P.O.W. Camp” by R. A. Radford, Economica, vol. 12, 1945 where cigarettes were the medium of exchange.

    Although Krugman’s favorite, the Washington DC Baby Sitting Coop story, involved sticky wages (scrip was defined in terms of hours of baby sitting) the stickiness was defined into the system so it smacks of price controls. Thus Explanation #2 still holds the most weight for me.

    Krugman likes the Washington Baby Sitting Coop story so much he has used it in two books. but the problem with that is that it had price controls (scrip was defined in terms of baby sitting).

  40. Gravatar of justanothereconomist justanothereconomist
    23. March 2011 at 22:20

    Scott-

    What causes money to affect NGDP? If your argument is that wages rise versus NGDP, then there’s two possibilities. If this is driven by a fall in NGDP, this is 100% tautological. If this is driven by a rise in nominal wages, then this argument doesn’t hold water, as wages are acyclical. So they can’t be driving the monetary channel.

    I’m unconvinced, but I’m unconvinced by many monetary channels (while remaining convinced that money matters outside of liquidity traps).

  41. Gravatar of edeast edeast
    24. March 2011 at 07:06

    dlr:

    In your two person example, why is a medium of account necessary? Can’t the two traders bounce offers off each other, or will the woodford model not allow it?

  42. Gravatar of Nick Rowe Nick Rowe
    24. March 2011 at 07:42

    dlr: Woodford’s model has sticky prices. Replace my 3 women with 3 firms. There’s a salon, which employs workers to cut hair. Etc. Assume all 3 firms have fixed prices (in the short run) but wages are flexible. Start in LR equilibrium, then hit it with a shock, so it goes into recession. If Woodford’s model allows the 3 firms to barter, his model is logically incoherent. The 3 firms are leaving gains from trade on the table in a recession. The owners of the 3 firms will barter their outputs, use some to pay wages in kind (real wages are flexible, since money wages are flexible), and consume the rest themselves.

  43. Gravatar of Scott Sumner Scott Sumner
    24. March 2011 at 08:20

    dlr, Thanks for that info, but I’ve always viewed the medium of account as “money,” so his economy is not moneyless in my book. And I think Nick’s still got a case, if the supply of MOA is too low to match demand.

    Mark, Was there unemployment in POW camps? Business cycles? Sticky wages and prices?

    Justanothereconomist. It’s easy to explain how monetary policy can affect NGDP. Almost every model except the PK model predicts that. I prefer the excess cash balance mechanism–aka the “hot potato”. And I am assuming that tight money causes NGDP to fall, not nominal wages to rise. I agree that nominal wages are fairly acyclical.

    The tautology you refer to is a common mistake–confusing NGDP and RGDP. A business cycle is a change in RGDP. RBC types would deny that changing NGDP through monetary policy would cause a change in RGDP. So where is the tautology? I’m trying to explain why nominal shocks have real effects. If nominal wages are sluggish, as you correctly imply, then that provides a natural transmission mechanism if a 10% fall in M2 causes NGDP to fall 10%. At given nominal wage rates companies will cut back on hours worked. Wages must fall with NGDP to prevent unemployment.

    Nick, I agree that Woodford’s thought experiment does not refute your transmission mechanism.

  44. Gravatar of dlr dlr
    24. March 2011 at 09:08

    edeast, I was using three parties in my example (worker, factory, barber) and the presumption is they could not barter without overcoming the usual costs, because the payment system is still reliant on common prices to operate. See also my response to Nick, next.

    Nick, but I am arguing that you can imagine a Woodford cashless model where these firms *cannot* barter any more than they can in a typical monetary economy. How do they accomplish this barter under my description? If the payment system is reliant on prices to operate, the firms will have as much trouble bartering there as they do now. The payment system needs prices — in a common unit — from the point of exchange to operate effectively, it is not capable of simply matching quantities (this doesn’t seem impossible to imagine). And the point of exchange must be between only two parties (or face the usual double coincidence costs of multi-party meet ups). So the system obviates the need to hold money (because once it has quantity AND price in common units information, it can then match people up), but does not eliminate the need to use a unit of account and to express demand and supply in terms of that unit of account in decentralized trading.

    Do you agree with Scott that your “transmission mechanism” would still work in a system with a Woodfordian Medium of Account, no medium of exchange, and no ability to easily switch to Barter? I would have guessed that you don’t.

  45. Gravatar of Mark A. Sadowski Mark A. Sadowski
    24. March 2011 at 09:59

    Scott wrote:
    “Mark, Was there unemployment in POW camps? Business cycles? Sticky wages and prices?”

    Absolutely. Here is an excerpt from the conclusion describing the effects of a shortage of both cigarettes and food (a crippling combination of both demand and supply side shocks that led to an “economic depression”). The amount of trade and services was highly dependent on the quantity of the medium of exchange (cigarettes) which you’ll especially note if you read the entire paper (it’s not long):

    “The economic organization described was both elaborate and smooth-working in the summer of 1944. Then came the August cuts and deflation. Prices fell, rallied with deliveries of cigarette parcels in September and December, and fell again. In January, 1945, supplies of Red Cross cigarettes ran out: and prices slumped still further: in February the supplies of food parcels were exhausted and the depression became a blizzard. Food, itself scarce, was almost given away in order to meet the non-monetary demand for cigarettes. Laundries ceased to operate, or worked for £s or RMk.s: food and cigarettes sold for fancy prices in £s, hitherto unheard of. The Restaurant was a memory and the BMk. a joke. The Shop was empty and the Exchange and Mart notices were full of unaccepted offers for cigarettes. Barter increased in volume, becoming a larger proportion of a smaller volume of trade. Thus, the first serious and prolonged food shortage in the writer’s experience, caused the price structure to change again, partly because German rations were not easily divisible. A margarine ration gradually sank in value until it exchanged directly for a treacle ration. Sugar slumped sadly. Only bread retained its value. Several thousand cigarettes, the capital of the Shop, were distributed without any noticeable effect. A few fractional parcel and cigarette issues, such as one-sixth of a parcel and twelve cigarettes each, led to monetary price recoveries and feverish trade, especially when they coincided with good news from the Western Front, but the general position remained unaltered.”

    And here’s a free web posting. Everybody who’s interested in money should read it. Anyone who thinks money doesn’t matter will be shocked into reality:

    http://www.albany.edu/~mirer/eco110/pow.html

    P.S. Radford was required reading in “Political Order and Change” at the University of Chicago in the 1980s.

  46. Gravatar of Nick Rowe Nick Rowe
    24. March 2011 at 11:00

    Scott: “Nick, I agree that Woodford’s thought experiment does not refute your transmission mechanism.”

    What I’m wondering though, is whether my thought experiment might refute Woodford’s transmission mechanism?

    dlr: I’m not really following you well (possibly my fault, not yours).

    We can imagine there is a unit of account, and prices are fixed in that unit of account, but people can barter without using money, provided that all barter exchanges respect the relative prices as fixed in that unit of account. So 1 haircut is worth 1 lb of Venus dust, and 1 massage is worth 1 lb of Venus dust, and 1 manicure is worth 1 lb of Venus dust. No Venus dust exists. Why can’t Woodford’s people barter their way out of a recession?

  47. Gravatar of Doc Merlin Doc Merlin
    24. March 2011 at 11:23

    @Nick Rowe:
    Depends on if there are objects that are highly substitutes for venus dust. In the case for the united states, its rather difficult as people who make things which are highly substitutable for USD are sent to jail (hello liberty dollars).
    Without these substitutes barter has high transaction costs and thus imposes its own adverse effect on trade.

  48. Gravatar of Nick Rowe Nick Rowe
    24. March 2011 at 11:36

    Doc: agreed. But, AFAIK, there are no transactions costs in Woodford’s model. So barter should be costless. Is it explicitly or implicitly ruled out? If not, why can’t Woodford’s people in the cashless economy barter their way out of a recession?

  49. Gravatar of justanothereconomist justanothereconomist
    24. March 2011 at 15:00

    Scott-

    OK, I see it a little better. What I don’t totally understand is why excess cash balances require sticky prices and wages. I don’t think it does.

    Let’s say a 20% increase in the money base increases nominal spending by 20%. (As in Robert Hall and the Monetary Transmission). If Real GDP is far below trend, most of this increase in NGDP will appear as increases in Q, and P will increase much less than 20%. What if RGDP is at trend? Then prices will rise by 20% in the long run. If RGDP is above trend, then prices will rise by more than 20%. etc. No stickiness required. Prices may not move instantaneously, but the sluggishness of prices plays no part in monetary transmission.

    Put it this way- do you think that in a frictionless world money wouldn’t matter? I don’t, so I don’t understand why stickiness is central to New Keynesian and other theories of money.

  50. Gravatar of The Numeraire The Numeraire
    24. March 2011 at 15:33

    Meltzer is partially correct, but so is Scott Sumner.

    Scott argues that NGDP is far below trend compared to previous recoveries, but without pro-growth fiscal policy higher NGDP will just be a function of higher prices — much like what the early 1990’s looked like (when fiscal policy was all zero-sum austerity).

    Obama caved in on the tax cut extension because of Republican pressure and now seem likely to reform the corporate tax, with lower marginal rates in exchange for having status quo companies drop some loopholes. Look at how many progressive types feel cheated by having supported Obama.

  51. Gravatar of James Oswald James Oswald
    25. March 2011 at 10:10

    “I think sticky wages are the key to money non-neutrality, but can’t prove it to those who use more conventional models.”
    I don’t think you have to; just marshal the facts! I can’t figure out how to embed it, but take a look at the following graph on the St. Louis Fed site:
    CPIAUCSL (CPI for all Urban Consumers, Seasonally Adjusted)
    GDP
    AHETPI (Average Hourly Earnings, Total Private Industries)
    All “percent change from a year ago”. I can email you the graph if you like.

    In October 2008, GDP and the CPI collapse, but wages don’t even budge until a few months later, when they barely react at all. Even the growth rate doesn’t drop below 2%. When a pattern is so clear in the data, I think the burden of proof should be on those who claim it’s not important enough to be in their models.

  52. Gravatar of James Oswald James Oswald
    25. March 2011 at 13:57

    @The Numeraire: “pro-growth fiscal policy higher NGDP will just be a function of higher prices”
    Fiscal stimulus works by raising spending. Monetary stimulus works by raising spending. If you want to claim they have different impacts on Y vs. P, you must have a theory to explain why. Scott will probably claim their impact is identical. I would say that if Congress is better at targetting unemployed resources, then fiscal stimulus will cause more Y than P shifts and vice versa. I think the market is better at allocating resources in general, so if anything, monetary stimulus is more likely to increase output more than prices in the short run. In the long run, it’s all prices.

  53. Gravatar of Scott Sumner Scott Sumner
    25. March 2011 at 18:23

    Mark, Thanks for the tip, I recall that paper, but never read it.

    Nick, It’s hard for me to comment on whether you refute his transmission mechanism, as I never could take it seriously (assuming it was interest rates.)

    How would he explain how monetary shocks affect NGDP in an economy with no financial system, just cash and goods?

    But to answer you question, I agree that with barter it should be possible to escape any recession not caused by real forces. But doesn’t Woodford assume sticky prices, which would seem to prevent barter, as one of the two sides would refuse to trade if the price “stuck” at the wrong level.

    justanothereconomist, We often observe 99% deflation, or 99.9% deflation, in one day. It’s called a currency reform, and it has no real effects at all. Those are the biggest demand contractions in world history–and no unemployment results. The only explanation I can see is that wages and prices are 100% flexible during a monetary reform–by government fiat. All contracts are legally adjust 100 to 1 or 1000 to one.

    Numeraire, You said;

    “Scott argues that NGDP is far below trend compared to previous recoveries, but without pro-growth fiscal policy higher NGDP will just be a function of higher prices “” much like what the early 1990’s looked like (when fiscal policy was all zero-sum austerity).”

    That’s factually inaccurate, inflation fell in the early 1990s–there was no stagflation. If NGDP rises quickly, so will RGDP. At least with 9% unemployment and wage growth at very low levels.

    James, Thanks, and the same thing happened in 1930, when NGDP and the WPI plunged, and wages barely budged.

  54. Gravatar of Morgan Warstler Morgan Warstler
    25. March 2011 at 18:40

    “But to answer you question, I agree that with barter it should be possible to escape any recession not caused by real forces. But doesn’t Woodford assume sticky prices, which would seem to prevent barter, as one of the two sides would refuse to trade if the price “stuck” at the wrong level.”

    Gah!

    I gave a good answer to this… dlr repeated it, and Nick skipped it.

    It isn’t real barter when we have a cash system in place.

    What we have is people willing to sell off their excess inventory at a discount as long as they get to buy up other people’s excess inventory at a discount.

    Logically, Nick’s entire argument says:

    What we have is bunch of deal seekers, who are more willing to sell their stuff at a discount (cut their own prices) IF THEY ARE SURE most everyone else is cutting their prices too.

    The barter deal presumes there’s a discounted something in it you want.

    And once again, the INTERNET is making it very easy to quickly find cheaper things that you want. This is a big giant effect. It isn’t a little thing. The Internet means there’s no such thing as a printed menu. It also quickly assures sellers they should LIQUIDATE and move one.

  55. Gravatar of ssumner ssumner
    26. March 2011 at 17:27

    Morgan, It’s hard to resolve abstract macro issues by looking at the real world. Even if the internet is making barter easier than before, it’s not all that easy in an absolute sense. It’s hard to even imagine an economy with no money.

Leave a Reply