Reply to Noahpinion and Andy Harless on NGDP/inflation

Here’s Noah criticizing NGDP:

Matt’s main point is that because inflation is mismeasured, real GDP (“RGDP”) is also mismeasured. Sure, that’s true (though of course NGDP itself is not measured with anything remotely approaching certainty). But that doesn’t mean NGDP is “better” than RGDP. An accurately measured number is not necessarily more useful than a less accurately measured number, if the latter is something we care about and the former is something we really don’t.

In terms of economic well-being, we do not care about NGDP. Or at least, we shouldn’t. To see this, take the example of Zimbabwe. In 2007, Zimbabwe’s NGDP grew at a rate of over 60,000%! No, that is not a typo. The “total quantity of spending” in Zimbabwe at the end of 2007 was more than 60,000 times what it was at the beginning of 2007. For real.

Anyone still think NGDP is the “actual thing”?

The problem with NGDP is that you can’t eat it.  It really is just numbers on a page. Normally, it is a reasonable approximation to a measurement of stuff we can eat. But when someone starts playing around with the numbers on the page, as in Zimbabwe’s hyperinflation in 2007, NGDP stops being any kind of measure of stuff we can eat.

Noah misses several key points.  Matt Yglesias and I don’t just think there’s mismeasurement, we think there’s no clear explanation of what we are trying to measure.  I’m a pragmatist (as is Matt), so I’d be willing to overlook that problem.  But it’s far worse, even a correctly measured CPI wouldn’t be measuring the thing we’d like it to measure–nominal shocks.  I still think NGDP is the real thing–it’s the proper indicator of nominal shocks.

Monetary policy has two objectives; nominal stability, and high and stable employment.  Most people think inflation is the right variable to measure nominal instability.  That’s flat out wrong, for a half dozen reasons I explain in this recent post, and won’t repeat here.  NGDP fluctuations cause the welfare losses that we (wrongly) associate with inflation.  Now obviously with Zimbabwe the two are almost indistinguishable and it makes little difference, but the point is that NGDP is still the better indicator, especially when inflation is relatively low.

Noah talks like NGDP does a poor job of proxying RGDP, which is the real thing that we are interested in.  But that’s not what NGDP is all about.  It’s a nominal shock indicator, not an indicator of living standards.  I’d be the first to admit that RGDP, with all its faults, is a better measure of living standards than NGDP.  But monetary policy doesn’t target living standards, the twin objectives are (should be) nominal stability and stable employment.  Prices and RGDP aren’t useful in developing optimal monetary policy.

Noah continued:

My point is that NGDP is not supposed to be a measure of how well the economy is doing. People who support NGDP targeting, like Scott Sumner, support it because they believe that NGDP is a good indicator of how much money we should print. Not because they think that NGDP is intrinsically good.

NGDP may be less indirectly measured, but RGDP is still the “actual thing.”

Low and stable NGDP growth is intrinsically good in the same sense that many (wrongly) believe low and stable inflation is intrinsically good.  It reduces nominal shocks.  The level of NGDP is unimportant, for the same reason that the level of prices is unimportant.  (I use “intrinsic” in the everyday sense of the word, arguably everything except utility is instrumental.)

But at least we agree on Radiohead.  An excellent band, but not a game changer.

Andy Harless agrees with some of my points, but argues that inflation is still needed for certain purposes.

I don’t think (and I doubt you do either) that economists should completely give up trying to measure the general price level or modeling it in the abstract. But we need to recognize that it becomes a messy concept once we take into account that there is more than one good. And it becomes an even messier concept once we take into account that the subjective “good” of any physical good doesn’t necessarily have a stable or simple relationship to the measurable quantity. And it becomes yet messier when we take into account that many goods are completely new and difficult to compare with pre-existing goods. And since the general price level is such a messy concept, it should be used with extreme caution when there is a much less messy concept available that can do most of the work that needs to be done.

It seems to me that there is precisely one area where inflation might be useful—living standards.  It’s hard to estimate the increase in RGDP without reference to some sort of inflation estimate.  But let’s consider whether it is actually useful in that area, not just potentially useful. 

The most recent debate over living standards relates to Tyler Cowen’s Great Stagnation hypothesis.  Cowen argues that growth in living standards slowed sharply after 1973, and that they are currently stagnating.  Others strongly disagree, arguing they continue to rise at a brisk rate.  So let’s suppose inflation was as useful as people claim.  How would the Great Stagnation debate be resolved?  Simple, just take per capita NGDP growth (which is fairly objective) and subtract the rate of inflation.  Voila, the debate is over.  But that’s not what I see in the blogosphere.  Instead, this is what I see:

Airplanes and cars!  . . .

                                         . . . the internet!

Home appliances! . . .

                                             . . . cell phones!

No more outhouses!

That doesn’t look very scientific to me.  I wonder why all these debaters don’t agree to simply dig up the NGDP and inflation data and do the calculations?  Why is that not seen as an answer to the question; “Are living standards improving as rapidly in recent decades, as during the post-war decades?”  And if the CPI can’t answer that question, what good is it?


Tags:

 
 
 

59 Responses to “Reply to Noahpinion and Andy Harless on NGDP/inflation”

  1. Gravatar of marcus nunes marcus nunes
    9. November 2011 at 10:00

    In this post:http://thefaintofheart.wordpress.com/2011/11/08/%E2%80%9Cweaving-a-story%E2%80%9D/
    I try to show that NGDP targeting is better at stabilizing the overall economy than IT. The fact that inflation has remained subdued almost everywhere in the face of NGDP instability in recent years, independently of the country being an inflation targeter, may simply be due to the manifestation of “PLOGS” (Persistently Large Output Gaps)

  2. Gravatar of Tommy Dorsett Tommy Dorsett
    9. November 2011 at 10:07

    From the early 1950s to the early 1970s, RGDP per capita compounded at a median growth rate of 2.6% per annum; from the early 1970s until 2007, RGDP per capita compounded at a median growth rate of 2.1%. Sure doesn’t seem like a Great Stagnation to me.

  3. Gravatar of Ram Ram
    9. November 2011 at 10:09

    I think the boldest claim you make here is that Radiohead isn’t a game changer. I’m disappointed Scott. I no longer support NGDP targeting…

  4. Gravatar of Ram Ram
    9. November 2011 at 10:16

    More substantively, I think the point to make here is that advocates of NGDP targeting don’t think, “the more NGDP the better.” Nor do they think NGDP is a reasonably good measure of the quality of public policy. Rather, they think that maintaing NGDP on a stable, rising path is the best way to minimize business cycle volatility, and thereby undesirably high inflation and unemployment. That’s it. That’s one very small part of our larger public policy objectives. Suggesting that maximizing NGDP is the only objective of public policy is to completely neglect what prominent advocates of NGDP targeting like yourself have actually said (over and over again).

  5. Gravatar of Silas Barta Silas Barta
    9. November 2011 at 10:31

    *sigh* … They don’t agree to just “dig up the NGDP and inflation numbers” because measuring inflation regresses back to the same question and forces you to answer the same questions! So bringing in inflation numbers will just recapitulate the debate they were already having!

    “Do we count newspapers as more expensive, or cheaper, since you can access a bunch of them online for free?” “Do we count entertainment as more expensive, or cheaper, since you can get a lot more e.g. music of higher quality and take it with you more places?”

    But then, *this* kind of inflation is going to have a different value than the kind you want to measure for purposes of monetary policy. Call the former QL_inflation and the latter MP_inflation. You can have low QL and high MP inflation (indeed, that’s where I think we are now), if the same experiences are getting cheaper (e.g. because of technology) while but monetary policy is making it so that your earnings buy you significantly less than you *could have* been able to buy had the CB not debased your savings.

    What’s worse is that central banking/inflation shills mix the two measures (like I showed here), and *only* count QL deflation if and when it lets them get away with MP inflation: “No, you don’t understand, computers getting cheaper is *good* deflation, and we don’t care about it when we come up with monetary policy … oh, I mean, except when we need to hide inflation when we’re bailing out banks.”

  6. Gravatar of Andy Harless Andy Harless
    9. November 2011 at 10:36

    If you want to model preferences (and presumably economists should model preferences if you think the Lucas Critique is important), it’s a lot easier to model them by allowing the price level to enter as a separate variable. If I’m making an intertemporal optimization decision, I care about how much real income I will have (RGDP) and how much my assets will be able to buy (prices). I don’t care directly about NGDP.

  7. Gravatar of Patrick R. Sullivan Patrick R. Sullivan
    9. November 2011 at 11:29

    Had the Zimbabwean CB been run by Sumnerians lo these many years, that country wouldn’t be better off today. Is that his argument?

  8. Gravatar of Tahtweasel Tahtweasel
    9. November 2011 at 11:45

    Zimbabwe’s Central Bank was printing money to try to cover long-term fiscal insolvency.

    Long-term fiscal insolvency leads to default or to collapse of the currency. There is no other way to end it. Zimbabwe chose the latter.

    I’m not sure you can be a Sumnerian in the context of a system where the fiscal policy has made financial collapse a certainty.

  9. Gravatar of Brian Brian
    9. November 2011 at 12:16

    Wow, Noah’s post is (for lack of a better term, and I don’t mean to be harsh) terrible.

  10. Gravatar of Becky Hargrove Becky Hargrove
    9. November 2011 at 12:22

    Brian,
    You’re right, I was wondering if someone was going to say something. Why did he give in to ego concerns (jealousy?) in uncertain financial times like the present?

  11. Gravatar of Keith Keith
    9. November 2011 at 12:25

    Scott,

    My primary question with NGDP targeting is what are the decision rules? Over what timeframe do you measure NGDP? What handles do you adjust in response and over what timeframe? When do you start? When do you stop? How much? Is it possible to develop a set of fixed decision rules that can be strictly followed?

  12. Gravatar of JimP JimP
    9. November 2011 at 12:40

    A clever euro idea:

    http://www.project-syndicate.org/commentary/james60/English

    Just let any euro member print – print their own currencies – that can circulate right beside the euro. They don’t have enough currency – so let them PRINT. The ECB wont print – but they will.

  13. Gravatar of Noah Noah
    9. November 2011 at 12:49

    Yes, I definitely understand that NGDP’s use is as a measure of nominal shocks – i.e., aggregate demand – not as a measure of well-being. I believe I said that in my post…

    My point is that most people out there don’t get this. And so all this talk of NGDP targeting, while it may be a step toward better monetary policy, seems to be accidentally tricking large swathes of the intelligent lay public into thinking that NGDP is a measure of living standards…

  14. Gravatar of JimP JimP
    9. November 2011 at 12:57

    Of course – if Italy did print some Lira now the government would have to agree to accept the new Lira for tax payments – and people who are owed Italian debt payments in euro might object.

    But all is fair in love and war.

  15. Gravatar of Adam Ozimek Adam Ozimek
    9. November 2011 at 14:00

    Scott,

    As a pragmatist surely you agree we need to measure living standards, right? And while we don’t have universal agreement on the “right” CPI, even conceptually, is there some other measure you’d prefer for that purpose? I think the pragmatist is left stuck with the CPI for cost of living measurement purposes.

  16. Gravatar of Benjamin Cole Benjamin Cole
    9. November 2011 at 14:36

    OT but fun:

    Evidently, the “Market Monetarism” entry into Wikipedia has succeeded, and is a permanent fixture. I don’t know who wrote what in the entry (I wrote some, edited some) but the final result not bad at all, and in fact is commendable.

    It is heartening to read the entry, in which one gets a sense at how rapidly market monetarism has entered the business mainstream.

    Next stop: The Fed. Now we go after the Big Banana to get the whole enchilada.

  17. Gravatar of Paul Andrews Paul Andrews
    9. November 2011 at 14:46

    “even a correctly measured CPI wouldn’t be measuring the thing we’d like it to measure-nominal shocks”

    So it has nothing to do with measuring currency debasement? Try telling that to holders of the currency.

  18. Gravatar of John John
    9. November 2011 at 14:55

    Scott,

    It’s an interesting point that poorly measured inflation makes RGDP look better. If you look at graphs of RGDP and CPI over 100 years, they both grow faster after the US left the gold exchange standard in 1971. I have a hard time believing that the US economy was as strong as the RGDP figures indicate after 1971.

    Even guys like Krugman say that the US economy hasn’t delivered for average Americans after 1971. Krugman says that it’s because of privatization and deregulation (yeah right). I say it’s because inflation is hurting the middle class while making growth look ok.

  19. Gravatar of John John
    9. November 2011 at 14:59

    Tahtweasel,

    I used to agree with you about long term fiscal insolvency ending in hyperinflation or default. I now think that there’s a third option, Japan. A country that issues it’s own currency can never become insolvent and hyperinflation doesn’t really happen to economies as long as anyone has confidence in them. People really have to start seeing the pieces of paper the state issues as worthless garbage before you can get a real currency collapse.

  20. Gravatar of Peter N Peter N
    9. November 2011 at 15:42

    Simple, just take per capita NGDP growth (which is fairly objective) and subtract the rate of inflation. Voila, the debate is over.

    Why is that not seen as an answer to the question; “Are living standards improving as rapidly in recent decades, as during the post-war decades?” And if the CPI can’t answer that question, what good is it?

    The CPI can be very bad used as part of calculating utility. There are all sorts of ways of increasing NGDP that don’t increase utility (The Chinese are currently conducting a massive uncontrolled experiment in this area).

    Do you want to consider leisure hours, real disposable income, purchase power equivalent disposable income, hours worked per household, median estimated real lifetime earnings? It would be fascinating to do all at once and compare. You would settle some disputes.

    In any case Cowen is wrong. What is happening is that we keep increasing the institutional barriers to change at the same time as we accelerate the rate of discovery. If we could just get half the stuff that we have out of the lab, the results would be amazing.

    Building robots is now a grade school project, IBM just simulated a cat brain. We’re finally giving psychology a biological basis. Photo-voltaic power has reached price crossover in the southwest. We can create 3D objects by printing in metal, plastic or ceramic. The cost of sequencing a human genome is around $1000 and falling. It cost billions and took from 1900 to 2003 to do the first one.

    Now if we could just do something about the patent office and the FDA and put a $500 an ear bounty on lobbyists…

  21. Gravatar of FXKLM FXKLM
    9. November 2011 at 15:42

    Noah,

    “My point is that most people out there don’t get this. And so all this talk of NGDP targeting, while it may be a step toward better monetary policy, seems to be accidentally tricking large swathes of the intelligent lay public into thinking that NGDP is a measure of living standards…”

    In all of the recent press on NGDP targeting, I don’t recall seeing anyone make that mistake.

  22. Gravatar of ssumner ssumner
    9. November 2011 at 18:12

    Marcus, Thanks for the link. That’s what I find wrong with IT as well, Inflation does not provide as good an indicator as NGDP.

    Tommy, Yes, the real GDP data is clear, so why don’t people believe it?

    Ram, I do like Radiohead, indeed they are one of my favorite modern bands. But I’m hopelessly stuck in the 1960s and 1970s.

    Plus I have lousy taste in music.

    I agree on the economics.

    Silas, Sounds like you agree that inflation numbers can’t be trusted.

    Andy, You said;

    “I don’t care directly about NGDP”

    You may be right, but that’s not clear to me. Suppose I learned that NGDP was going to rise at 20% a year over the next few decades. I’d want to consume more now, borrow against my future income. And I’d want to do that regardless of whether the fast NGDP growth was inflation or real growth. Tell me what I am missing?

    Also assume that utility depends on how you are doing compared to your neighbors.

    Patrick, No, I think he understands that I favor low and stable NGDP growth, not “the more the merrier”, but I think he misses part of the argument for why NGDP is superior to inflation.

    Tahtweasal, That’s right.

    Brian and Becky, I should say that when I’ve read him before, he’s seemed like a smart guy. It’s hard to jump into an esoteric debate over NGDP, and understand all the nuances of the issues. I see many “outsiders” having that problem, now that the subject is suddenly becoming popular.

    Keith, There are many possibilities—I’ve even suggested the government try to construct daily estimates. But I think the most reasonable policy would be either a one or two year forward target of NGDP. It might be daily, monthly or quarterly.

    JimP, It might help, but there are all sorts of potential problems with how contracts are going to be denominated. I’ll keep an open mind, and try to think about it a bit more.

    Noah, OK, but what threw me off is your criticism of Yglesias. He certainly wouldn’t disagree with anything you say here, so I’m not clear why you had a problem with Yglesias’s post.

    Adam, You said;

    “As a pragmatist surely you agree we need to measure living standards, right?”

    You need to tell me why, before I decide whether there is a better metric than inflation. I recently argued the CPI isn’t the right one for indexing pensions, for instance.

    I do concede that RGDP is probably better than NGDP for certain purposes. But what specific “need” are you referring to?

    To summarize my views:

    1. Inflation isn’t a “thing” out there in nature, to be measured more or less accurately.
    2. The CPI may be useful for a small number of purposes, but that has yet to be shown.

    Ben, Thanks for contributing to that entry. I’m not sure who did it, nor do I know who did the Scott Sumner entry.

    Paul, You said;

    “So it has nothing to do with measuring currency debasement? Try telling that to holders of the currency.”

    Currency demand is better measured by the opportunity cost of holding currency (the nominal interest rate) rather than the inflation rate.

    John, Inflation definitely picked up (using any index), but I believe that RGDP growth slowed after 1973.

    Peter, I think Cowen is right. Living standards grew way faster during my grandma’s life than during my life. My life has seen computers and cell phones added. My grandma went from candles and outhouses and horses to the US landing on the moon. We will never again see anything like that rate of progress.

  23. Gravatar of JJ JJ
    9. November 2011 at 19:15

    If everyone knew there would be 20% NGDP growth, asset prices would soar. Land, buildings, trees, oil in the ground and long duration assets in general.

  24. Gravatar of johnleemk johnleemk
    9. November 2011 at 19:27

    Ben/Scott:

    Most Wikipedia articles are co-written. I wrote a fair share of the Sumner article, but as far as I can tell a bunch of people co-authored the market monetarism article. Ben, I argued quite strenuously for keeping the article, I didn’t expect to win at first, but Krugman’s post on market monetarism saved the day: http://en.wikipedia.org/wiki/Wikipedia:Articles_for_deletion/Market_monetarism

    On an unrelated note, I wonder what Morgan makes of Rick Perry after tonight’s debate.

  25. Gravatar of Lorenzo from Oz Lorenzo from Oz
    9. November 2011 at 20:12

    Tyler Cowen’s “great stagnation” argument seems very American-centric. We haven’t noticed any stagnation Downunder. And it is not as if we have much in the way of “technological catchup” to do.

    There is an argument about the distribution of gains in the US economy, but that seems a much more fraught argument.

  26. Gravatar of Lorenzo from Oz Lorenzo from Oz
    9. November 2011 at 20:14

    More fundamentally, predicting future knowledge seems like the ultimate oxymoron.

  27. Gravatar of Paul Andrews Paul Andrews
    10. November 2011 at 00:34

    @Scott:

    I said: “So it has nothing to do with measuring currency debasement? Try telling that to holders of the currency.”

    You replied: “Currency demand is better measured by the opportunity cost of holding currency (the nominal interest rate) rather than the inflation rate.”

    Did you misread my comment? If not, in what way is currency demand the same concept as currency debasement?

  28. Gravatar of Paul Andrews Paul Andrews
    10. November 2011 at 00:50

    @Scott:

    You said: “You may be right, but that’s not clear to me. Suppose I learned that NGDP was going to rise at 20% a year over the next few decades. I’d want to consume more now, borrow against my future income. And I’d want to do that regardless of whether the fast NGDP growth was inflation or real growth. Tell me what I am missing?”

    20% inflation: If interest rate is lower than 20%, borrow like crazy and invest in non-productive assets that will hold their real value, and increase in nominal value. My debt will magically be inflated away. If interest rate is higher than 20%, pinch myself (not really compatible with decades of 20% inflation).

    20% real growth: Get on board the real growth train. Invest in productive assets, work hard, or get left behind.

    The two situations are completely and utterly different.

    It seems that NGDP growth targeting can only makes sense when you ignore most of the important stuff about the way humans behave.

    There’s no point encouraging borrowing for borrowing’s sake. What matters is how the borrowing is invested. Poor investments need to be curtailed, not ameliorated by inflation.

  29. Gravatar of Peter N Peter N
    10. November 2011 at 02:38

    “My grandma went from candles and outhouses and horses to the US landing on the moon. We will never again see anything like that rate of progress.”

    And where did she live?

    My great grandmother had indoor plumbing and electric lighting (introduced in 1887 in Brooklyn. Her mother only had gas, and her grandmother, I believe, used oil lamps, but she lived upstate. If she’d lived in Manhattan, she could have had gas light). She and her husband used to drive across the (electrically lighted) Brooklyn Bridge in their carriage to dine at Delmonicos in the 1890s.

    OTOH our country house in upstate NY still had an outhouse into the late 1950s, though we had 1 indoor toilet installed around 1949.

    Flush toilets were invented in the 26th century BC. They were common throughout the Roman empire. The Tremont Hotel in Boston was the first with indoor plumbing in 1829. The tenement act of the early 1900s in NYC required a flush toilet in every apartment. Earlier “old law” tenements usually had one toilet per floor (I’ve lived in one).

    And how many people have landed on the moon since? It cost $100 billion (2010$) about 35% of NASA spending over 10 years. That’s with little regard to or understanding of the risks. It would take maybe $300 billion to do it with today’s standards. Write your congressperson. OTOH NASA has done amazing things. Rovers on Mars, Cassini, weather satellites, Space telescopes.

    The more stuff you have, the more you have to destroy to build new stuff, unless you’re talking about gadgets like the IPOD. That’s largely why we have IPADs but not high speed rail between Boston and Washington. The cost of the rights of way is appalling, since the route runs right through all the cities along the way.

    It’s hard to replace good enough with better unless it’s painless. In 5 to 10 years LED lighting will wipe out the incandescent bulb. Now it’s 20 times as expensive, lasts 40 times as long, and uses 10% of the power. Payback for me is under 2 years and falling fast.

    With more infrastructure, more bureaucracy, and better communications (so it’s harder to sneak something by those who would oppose it), it takes longer to get innovation adopted, and innovators concentrate on areas where adoption is easier. Most of the life changing inventions of the 20th century were invented in the 19th or even earlier. A 50 year or more time to adopt was fairly typical.

  30. Gravatar of dwb dwb
    10. November 2011 at 05:58

    some of these blogosphere posts on NGDP have the flavor of resignation that its a “done deal” and seem to expect the Fed to change to NGDP targeting. things that make you go hmmm.

  31. Gravatar of JimP JimP
    10. November 2011 at 06:08

    Scott

    As the guy says about silver – use the local currency for local stuff – wages primarily – and the euro for international things. Nor would the government have to accept the local currency for taxes – as there would always be an exchange market from the local into euros. The main thing is to break the currency monopoly of an institution that refuses to supply enough of it. One could even imagine a backed – up local currency issued by banks – go back to bimetallism and free banking as well.

  32. Gravatar of Ryan Ryan
    10. November 2011 at 07:52

    Prof. Sumner, you say: “But it’s far worse, even a correctly measured CPI wouldn’t be measuring the thing we’d like it to measure-nominal shocks. I still think NGDP is the real thing-it’s the proper indicator of nominal shocks.”

    You previously indicated that an increase in Aggregate Demand occurs when output and NGDP both rise.

    So, two questions:

    1 – Do you accept my assertion that a rise in NGDP with no corresponding rise in output is strictly the bad kind of inflation, ceteris paribus?

    2 – When there is no change in output and an economy experiences a “nominal shock,” what is literally happening in the economy? What is that which would cause a “nominal shock” but would not also cause a real shock?

  33. Gravatar of Doc Merlin Doc Merlin
    10. November 2011 at 11:10

    NGDP targeting wouldn’t allow the central bank to weasel its way out of things, and wouldn’t allow the politicians to blame the central bank for economic problems.

    This alone discourages me to the idea that its possible for it to come to pass.

  34. Gravatar of ssumner ssumner
    10. November 2011 at 19:01

    JJ, Not if nominal interest rates were 20%.

    Johnleemk, Thanks for your great work on Wikipedia.

    Lorenzo, I think his argument is partly based on recent trends, not just the future. Also recall that growth has lots of momentum. The discovery of electric power and internal combustion engines powered growth for many decades. He doesn’t see new discoveries than can have that effect. I don’t either.

    But I agree that Australia’s doing well.

    Paul, My point is that currency holders feel bad because NGDP is rising fast, not because prices are rising fast. NGDP is a better metric for their suffering.

    On your second comment, my point is that either way cash holders can do much better investing their money, and returns will be high.

    Obviously the types of investments will differ, but that doesn’t relate the Andy’s thought experiment.

    Peter, You said;

    “My great grandmother had indoor plumbing and electric lighting (introduced in 1887 in Brooklyn. Her mother only had gas, and her grandmother, I believe, used oil lamps, but she lived upstate. If she’d lived in Manhattan, she could have had gas light). She and her husband used to drive across the (electrically lighted) Brooklyn Bridge in their carriage to dine at Delmonicos in the 1890s”

    Not many Americans had electric lights and indoor plumbing in the 1890s–certainly not my grandma in Wisconsin.

    Again, My grandma saw amazing changes in all areas of her life. I’ve seen it in one–computers.

    In the 1890s the airplane hadn’t been invented. When she died in 1969 the Boeing 747 was already flying. It’s still basically state of the art. Progress has obviously slowed sharply.

    dwb, I still think it unlikely in the short run, but possible in the next recession.

    JimP, The issue for me is two-fold. How is it not counterfeiting? And if not (and I presume it isn’t) then don’t you have the big inconvenience of two units of account? Maybe it’s worth that inconvenience, given how bad things are.

    But if the powers-that-be would allow this, why don’t they just do the right thing in the first place?

    Ryan, I see a nominal shock as being something that primarily throws the labor market out of equilibrium. Although the debt market is also an issue.

    If NGDP grows 5%, I’m happy with monetary policy regardless of whether it’s inflation or RGDP growth. But if it’s all inflation, I’m very unhappy with the government’s supply-side policies. See my recent “Bush league” post.

    I’m not sure I answered your question, if not ask it a different way.

    Doc Merlin, Good argument–but make it more positive. That’s why we must pass it.

  35. Gravatar of Paul Andrews Paul Andrews
    10. November 2011 at 19:50

    @Scott:

    “Paul, My point is that currency holders feel bad because NGDP is rising fast, not because prices are rising fast. NGDP is a better metric for their suffering.”

    The objective is not simply to make currency holders feel bad.

    The difference is that when NGDP is rising fast but prices are staying steady, the currency holder will “feel bad” because others are doing better, but at least his purchasing power will stay steady.

    If NGDP is rising fast and prices are rising at the same rate, the currency holder will “feel bad” because his purchasing power is being eroded. And the non-currency holders are merely treading water.

    It’s better for the economy as a whole if currency holders feel bad for the first reason (because everyone else is being more productive than the currency holder), than for the second.

    “On your second comment, my point is that either way cash holders can do much better investing their money, and returns will be high.”

    Yes *nominal* returns will be high in both cases, but real returns will not. Increasing real returns is a much more sensible objective than increasing nominal returns.

  36. Gravatar of Paul Andrews Paul Andrews
    10. November 2011 at 19:55

    “If NGDP grows 5%, I’m happy with monetary policy regardless of whether it’s inflation or RGDP growth. But if it’s all inflation, I’m very unhappy with the government’s supply-side policies. See my recent “Bush league” post.”

    Do you admit the possibility that the “good” monetary policy in this case enables, at least partly, the “bad” fiscal policy, by allowing the government to borrow a portion of its funds interest-free? (Assuming monetary policy involves the purchase of government debt by the Fed).

  37. Gravatar of Ryan Ryan
    11. November 2011 at 05:01

    Prof. Sumner,

    Is there anything that would throw the labor market out of equilibrium that would not *also* have a significant impact on RGDP?

    * If yes, how is this possible?

    * If no, why is this a “nominal” shock?

  38. Gravatar of ssumner ssumner
    12. November 2011 at 14:08

    Paul, Economists think in terms of opportunity cost, it’s the only cost that matters.

    Paul, You said;

    “Do you admit the possibility that the “good” monetary policy in this case enables, at least partly, the “bad” fiscal policy, by allowing the government to borrow a portion of its funds interest-free? (Assuming monetary policy involves the purchase of government debt by the Fed).”

    No, with steady 5% NGDP growth interest rates would be much higher.

    Ryan, A nominal shock does have real effects, that’s exactly the point. But the effects only occur because wages and prices are sticky. A real shock affects output even if wages and prices adjust instantly.

  39. Gravatar of Paul Andrews Paul Andrews
    12. November 2011 at 16:35

    @Scott:

    “Paul, Economists think in terms of opportunity cost, it’s the only cost that matters.”

    So economists are not interested in ensuring an efficient free market by ensuring that non-productive investments are curtailed?

    “No, with steady 5% NGDP growth interest rates would be much higher.”

    This is not the case if the interest rate is being kept artificially low in order to increase inflation so as to maintain the 5% NGDP target.

  40. Gravatar of Scott Sumner Scott Sumner
    13. November 2011 at 11:05

    Paul, You get stable free markets with steady NGDP growth. That’s how you allocate resources efficiently. I oppose all governemtn attempts to manipulate interest rates. They should be set in the market. Tight money produces 0% rates.

  41. Gravatar of Paul Andrews Paul Andrews
    13. November 2011 at 23:53

    Scott, “You get stable free markets with steady NGDP growth” – this is begging the question.

  42. Gravatar of Ryan Ryan
    14. November 2011 at 09:20

    Prof. Sumner,

    So then a 5% NGDP growth policy is a nominal shock unless wages increase 5% annually?

    Where can I find one of these jobs that offers me a 5% annual raise?

  43. Gravatar of Scott Sumner Scott Sumner
    14. November 2011 at 18:18

    Paul, How is it begging the question?

    Ryan, I didn’t say 5% NGDP growth was a nominal shock, and I didn’t say it would produce 5% wage growth.

  44. Gravatar of Paul Andrews Paul Andrews
    15. November 2011 at 01:30

    Scott,

    Andy said: “I don’t care directly about NGDP”

    You said: “You may be right, but that’s not clear to me. Suppose I learned that NGDP was going to rise at 20% a year over the next few decades. I’d want to consume more now, borrow against my future income. And I’d want to do that regardless of whether the fast NGDP growth was inflation or real growth. Tell me what I am missing?”

    I said: “20% inflation: If interest rate is lower than 20%, borrow like crazy and invest in non-productive assets that will hold their real value, and increase in nominal value. My debt will magically be inflated away. If interest rate is higher than 20%, pinch myself (not really compatible with decades of 20% inflation).

    20% real growth: Get on board the real growth train. Invest in productive assets, work hard, or get left behind.

    The two situations are completely and utterly different.

    It seems that NGDP growth targeting can only makes sense when you ignore most of the important stuff about the way humans behave.

    There’s no point encouraging borrowing for borrowing’s sake. What matters is how the borrowing is invested. Poor investments need to be curtailed, not ameliorated by inflation.”

    You said: “Paul, My point is that currency holders feel bad because NGDP is rising fast, not because prices are rising fast. NGDP is a better metric for their suffering.”

    I said: “The objective is not simply to make currency holders feel bad. (and subsequent remarks)”

    You said: “Paul, Economists think in terms of opportunity cost, it’s the only cost that matters.”

    I said: “So economists are not interested in ensuring an efficient free market by ensuring that non-productive investments are curtailed?”

    You said: “You get stable free markets with steady NGDP growth”

    I said: “this is begging the question.”

    You said: “Paul, How is it begging the question?”

    What I mean is that your point “You get stable free markets with steady NGDP growth” seems to merely be restating your original premise, which I am arguing can lead to instability due to malinvestment.

  45. Gravatar of Ryan Ryan
    15. November 2011 at 05:49

    Prof. Sumner,

    You certainly didn’t say NGDP growth was a nominal shock – but based on how you have defined a nominal shock, I can only conclude that it is.

    You said that a nominal shock was a shock that only has real effects because wages and prices are sticky. NGDP changes with no corresponding change in output. (Also your definition of an AD shock.)

    Setting aside that this looks to me to be very close to “reasoning from a price (NGDP) change,” a 5% NGDP growth policy is a nominal shock to me because my wages are sticky. I spend more without any corresponding increase in my own output (or income).

    Given that my wage is not at all tied to inflation and is therefore sticky, isn’t it true that a perpetual NGDP growth policy always and forever hurts me?

  46. Gravatar of ssumner ssumner
    15. November 2011 at 13:13

    Paul, I don’t think the Fed should try to allocate resources between various investment projects, rather I think they should aim for low inflaiton on average (actually NGDP grwoth, but most people prefer to talk in terms of inflation) and then a policy that doesn’t needlessly make the economy more unstable. The economy will always have cycles, but if NGDP fluctuates those cycles will be needelessly enlarged. For me the key macro problem (market failure) is sticky wages. Interest rates aren’t sticky (with NGDP targeting the Fed no longer targets rates) so I’d rather let the free market allocate investment, and focus on avoiding unneeded unemployment due to sticky wages. You do that with stable NGDP growth.

    Ryan, If we had a 5% NGDP targeting regime, then presumably you’d negotiate your wage contract on expectations of 5% NGDP growth. But that’s not the same as claiming that you’d negotiate a 5% wage gain, real wages will vary–plus population will grow.

  47. Gravatar of Paul Andrews Paul Andrews
    16. November 2011 at 05:14

    We had stable NGDP growth for a number of years, coincident with highly inefficient resource allocation, resulting in a financial crisis and high unemployment.

  48. Gravatar of Ryan Ryan
    16. November 2011 at 06:28

    Prof Sumner, if my wage contract allowed for a 5% nominal increase per year, than my wages wouldn’t be sticky at all.

    As I said two comments ago: Where can I find one of these jobs that offers me a 5% salary increase per year? In my experience, inflation-adjusted salaries are only available to employees in government-funded jobs or heavily unionized industries. Everywhere else, raises are tied to job performance, not “expectations of future inflation.” You can certainly disagree with my experience, but then it becomes a game of “Who are you going to believe, Ryan – Scott Sumner, or your lying eyes?”

    NGDP targeting won’t hurt me, so long as I build public-sector employment into my expectations. I can take a government job with inflation-adjusted wages any time I want to, but this seems like a rather silly economic policy. I should hardly be forced to take a government or union job just to protect myself against inflationary monetary policy. Or do you disagree?

  49. Gravatar of Scott Sumner Scott Sumner
    16. November 2011 at 16:59

    Paul, We had stable NGDP growth coinciding with global warming, and the Patriots winning three Super Bowls. Are they connected? No one would claim NGDP targeting solves real allocation of resources problems. It’s not the Fed’s job to focus on an industry making up 6% of GDP (housing) and tailor policy to that industry. If there are abuses, change the regulatory structure.

    Ryan, You said;

    “Prof Sumner, if my wage contract allowed for a 5% nominal increase per year, than my wages wouldn’t be sticky at all.”

    I don’t think you understand what sticky wages are. They would be sticky. You seem to think sticky wages means constant wages–not so.

    Thus the rest of your comment is off target, as you are assuming I’m claiming something that I am not claiming.

    BTW, there is no sense in arguing that wages aren’t sticky, the empirical evidence is stronger than virtually any other proposition in economics, or any other science.

  50. Gravatar of Paul Andrews Paul Andrews
    16. November 2011 at 21:42

    Scott,

    You said: “I’d rather let the free market allocate investment, and focus on avoiding unneeded unemployment due to sticky wages. You do that with stable NGDP growth.”

    We had a long period of stable NGDP growth that resulted in high unemployment.

    Therefore you cannot avoid unemployment simply through stable NGDP growth.

    Re: sticky wages: I don’t think anyone can reasonably claim that wages are not sticky. The question is, how sticky? And does loose monetary policy unintentionally make them more sticky, thereby preventing the market from clearing and exacerbating the problem it sets out to resolve?

  51. Gravatar of Ryan Ryan
    17. November 2011 at 06:41

    Prof. Sumner, that was a bit of a straw-man, so I’ll give you another shot at it.

    My assumptions (which are facts in my life):
    1 – My salary does not increase by anything close to 5% per year.
    2 – You prefer a world of ~5% NGDP targeting
    3 – In such a world, NGDP would increase by ~5% per year.

    My question:
    Why does this gap between inflation and my salary not hurt me in the long run?

    Will you answer that question, or will you wave it away by suggesting that what I am asking for is “constant real wages”? I get what you’re saying. I do.

    I am making the point that a world of perpetual inflation is a world in which my real income is perpetually diminishing. That sort of works against your growth hypothesis.

    Let’s accept the tenet of sticky wages and assume my expectations are spot-on. There still exist periods in which I am unable to respond to my own inflation expectations (if there were not, wages would not be sticky). How will I ever make up the difference between the inflation I know/expect and the income I am capable of earning?

    Now I understand the response to this: You might suggest that “on average” there is growth, i.e. that the income growth for the beneficiaries of venture capital is a greater total sum than the losses accrued by the rest of us… Okay, but why should “the rest of us” go for that?

    In other words, why should I care about a policy that promotes growth to people other than myself, at significant loss to myself?

    Give me a practical reason why I should go for that.

  52. Gravatar of Scott Sumner Scott Sumner
    17. November 2011 at 19:13

    Paul, You said: “We had a long period of stable NGDP growth that resulted in high unemployment.”

    Actually, unemployment only rose sharply in 2008-09, when NGDP fell.

    Ryan, You said;

    “Why does this gap between inflation and my salary not hurt me in the long run?”

    You are confusing NGDP with inflation.

    In addition, monetary policy has no effect on real wages in the long run, so I don’t understand your point.

  53. Gravatar of Paul Andrews Paul Andrews
    17. November 2011 at 21:15

    “Actually, unemployment only rose sharply in 2008-09, when NGDP fell.”

    When unemployment rises and NGDP falls, you claim that the fall in NGDP was the cause of the rise in unemployment.

    If unemployment rose and NGDP stayed stable, you would claim that the stable NGDP stopped unemployment from rising further.

    What set of empirical facts would prove your theory wrong?

  54. Gravatar of Ryan Ryan
    18. November 2011 at 07:58

    Oh, brother.

  55. Gravatar of Scott Sumner Scott Sumner
    18. November 2011 at 19:04

    Paul, I don’t claim all movements in U are caused by NGDP. But I do think that most dramatic moves in U are. If sudden and large changes in U were not associated with sudden changes in NGDP growth, that would greatly weaken my argument.

  56. Gravatar of Paul Andrews Paul Andrews
    19. November 2011 at 00:48

    Surely if NGDP targeting had been used in 2008, and was successful despite a sudden drop in U, this would have created exactly those circumstances.

  57. Gravatar of Scott Sumner Scott Sumner
    19. November 2011 at 18:40

    Paul, I don’t follow, why would a sudden drop in U be bad?

  58. Gravatar of Paul Andrews Paul Andrews
    20. November 2011 at 14:42

    Sorry I meant a sudden increase in U.

  59. Gravatar of Scott Sumner Scott Sumner
    21. November 2011 at 14:12

    Paul, Yes, that’s possible, but I think U would have gone up much less.

Leave a Reply