The NGDP moment

Back in 2008 you rarely heard economists talk about nominal GDP.  Now it’s the hot new idea in monetary policy targeting. But that’s not all, we are also seeing more interest in nominal GDP-linked assets.  Most famously, Robert Shiller has proposed “trills,” which would be assets with a value linked to one trillionth of nominal GDP. They could be used to hedge against aggregate nominal income risk. I’ve proposed nominal GDP futures markets, which are probably more accurately termed “prediction markets.” And now we have the new Greek government talking about NGDP-linked government debt:

Greece’s radical new government unveiled proposals on Monday for ending the confrontation with its creditors by swapping outstanding debt for new growth-linked bonds, running a permanent budget surplus and targeting wealthy tax-evaders.

Yanis Varoufakis, the new finance minister, outlined the plan in the wake of a dramatic week in which the government’s first moves rattled its eurozone partners and rekindled fears about the country’s chances of staying in the currency union.

After meeting Mr Varoufakis in London, George Osborne, the UK chancellor of the exchequer, described the stand-off between Greece and the eurozone as the “greatest risk to the global economy”.

Attempting to sound an emollient note, Mr Varoufakis told the Financial Times the government would no longer call for a headline write-off of Greece’s €315bn foreign debt. Rather it would request a “menu of debt swaps” to ease the burden, including two types of new bonds.

The first type, indexed to nominal economic growth, would replace European rescue loans, and the second, which he termed “perpetual bonds”, would replace European Central Bank-owned Greek bonds.

.  .  .

“What I’ll say to our partners is that we are putting together a combination of a primary budget surplus and a reform agenda,” Mr Varoufakis, a leftwing academic economist and prolific blogger, said.

A former blogger becomes a finance minister and proposes NGDP-linked bonds. What a crazy world we live in!

PS.  I have a post on Greece over at Econlog




42 Responses to “The NGDP moment”

  1. Gravatar of Nick Rowe Nick Rowe
    2. February 2015 at 18:57

    Dunno though. Who issues the statistics on Greek NGDP? The Greeks?

  2. Gravatar of ChacoKevy ChacoKevy
    2. February 2015 at 18:57

    For those (mainly me) who have trouble keeping names straight – Mr. Varoufakis is the man who took the position of Economist in Residence at Valve in 2012. Valve is operated by Gabe Newell. Mr. Newell was one of the big naames to pledge support to Mr. Sumner’s NGDP futures market project last year.
    Mr. Varoufakis had a few blog posts at Valve, too. His thoughts on digital economies are here:

  3. Gravatar of ssumner ssumner
    2. February 2015 at 19:15

    Nick, The US and UK governments issue the CPI data used in TIPS. Surely you are not suggesting that Greek governments are less honest than the US or UK government?

    If you aren’t suggesting it, I will. 🙂

    Very good point that I overlooked.

    ChacoKevy, I keep wondering if he was the Valve employee I spoke with over the phone a few years back.

  4. Gravatar of ChacoKevy ChacoKevy
    2. February 2015 at 19:35

    I don’t know anything about Valve’s ethnic composition, but I’m willing to guess they don’t have TWO Greek economists. Do you remember the individual having an accent?

  5. Gravatar of SG SG
    2. February 2015 at 19:48

    Would it really be easy for the greeks to cook the books on NGDP? As Scott has pointed out before, NGDP is the actual thing that we can measure with some precision – it’s RGDP and inflation that are both arbitrary and amorphous concepts that would be easy for a government to manipulate.

  6. Gravatar of Nick Rowe Nick Rowe
    2. February 2015 at 20:02

    SG: Yes. They did some dodgy accounting on the budget deficit to get into the Euro in the first place.

    There’s the whole underground economy for starters. Plus, do you count something as an intermediate good (not included in NGDP) or as an investment good (included in NGDP) if it depreciates quickly? Is that a government transfer payment (excluded) or payment for a service (included)? Etc.

    These ambiguities don’t matter all that much for NGDP targeting, as long as the statistic is calculated in a reasonably consistent way over time. But give a government a big incentive to understate NGDP, and a compliant national bureau of statistics, and watch what happens.

    It’s like me issuing shares in my income, and then you having to trust my calculations of what my income is.

  7. Gravatar of Ray Lopez Ray Lopez
    2. February 2015 at 21:57

    Geez Louise, this is not rocket science people. I feel like I’m reading some version of the Keystone Kops (“I don’t know anything about Valve’s ethnic composition, but I’m willing to guess they don’t have TWO Greek economists. Do you remember the individual having an accent?” and “ChacoKevy, I keep wondering if he was the Valve employee I spoke with over the phone a few years back.” LOL! Dumber and Dumber).

    Let me explain it to you in simple words even a person of 80 IQ can understand: Yanis Varoufakis –who has Marxist leanings– is a fan of Sumner’s blog; he’s stated he reads it. He was instrumental in getting his employer to throw money (away) to Scott’s NGDP prediction market. Yanis sees NGDPLT as a backdoor but seemingly respectable way to print money. As anybody who knows Greek politics knows (and I do, I am a citizen), the Greek government will lie about statistics (Google Costas Simitis and Euro) when it’s in their interest to do so, as Nick Rowe implies. So their targeting NGDP linked bonds will be a form of fraudulent financial repression.

    BTW, it’s a mistake for Yanis to throw away the one big card the Greeks have, which is the threat of default. I suspect he will be, around the summer of this year, forced to repudiate this imprudent move, when the Germans and French start making more demands than the Greeks can deliver. Much better to keep the threat of default–which BTW would do more good than harm long term, Argentina, Russia being exemplars–close to the vest to be used as an ace in the sleeve.

  8. Gravatar of Ben J Ben J
    2. February 2015 at 22:54


    You are pathetically stupid. Remind everyone what currency the former valve economist will be able to print?

  9. Gravatar of Ray Lopez Ray Lopez
    2. February 2015 at 23:33

    @Ben J–you are dim. He will print nothing, but his bank will print, with the permission of the Troika, in Euros. And the key point is that the “NGDP” statistic, as Rowe points out, will come from Greece, so that’s how GRE prints money fraudulently. Got it now?

  10. Gravatar of Lars Christensen Lars Christensen
    3. February 2015 at 00:54


    It should be said that it is not totally new – Greece actually back in 2012 issued (N)GDP as part of its latest debt deal.

    That said, it is a great idea and obviously the EU should support the Greek government’s efforts in moving in this direction.

  11. Gravatar of Blue Eyes Blue Eyes
    3. February 2015 at 04:47

    With regard to “trills”, don’t we already have plenty of such instruments? For example, real estate tends to broadly track NGDP (especially in the UK where we don’t ever build enough homes), because rents follow nominal incomes.

  12. Gravatar of Gdp-linked loans and loan restructuring for Greece Gdp-linked loans and loan restructuring for Greece
    3. February 2015 at 04:48

    […] Here are older proposals for gdp-indexed bonds (pdf).  Scott Sumner comments on the new Greek proposal. […]

  13. Gravatar of Blue Eyes Blue Eyes
    3. February 2015 at 04:50

    Nick Rowe “give a government a big incentive to understate NGDP, and a compliant national bureau of statistics, and watch what happens”

    This is what happened when Gordon Brown’s fiscal rules started to look more and more difficult to achieve. He couldn’t be seen to ditch the rules (although everyone at the time knew that is what he was doing) so he reclassified practically every bit of public spending as capital investment. Bazinga!

  14. Gravatar of Gdp-linked loans and loan restructuring for Greece – Freedom's Floodgates Gdp-linked loans and loan restructuring for Greece - Freedom's Floodgates
    3. February 2015 at 04:56

    […] Here are older proposals for gdp-indexed bonds (pdf). Scott Sumner comments on the new Greek proposal. […]

  15. Gravatar of Matt McOsker Matt McOsker
    3. February 2015 at 05:49

    While I like the Greek bond ideas ,including NGDP linked bonds, running surpluses and chasing wealthy evaders will not support their GDP growth. They need to be able to run deficits when needed, deficits of the proper size to offset any lack in private borrowing or trade deficits.

  16. Gravatar of Land Leaser Land Leaser
    3. February 2015 at 05:49

    Curious if this idea will be up for discussion at some point:

  17. Gravatar of Matt McOsker Matt McOsker
    3. February 2015 at 05:58

    Ray, Greece is no longer a fiat currency they cannot “print” euros. They are more akin to a U.S. state than a sovereign currency issuer.

  18. Gravatar of John Thacker John Thacker
    3. February 2015 at 06:13

    Argentina also offered GDP-linked bonds in 2002, IIRC. I don’t know all the details, but I know that there have been a lot of allegations of Argentina falsifying the GDP statistics as a result. Then again, they haven’t treated regular bond investors well either.

  19. Gravatar of Ray Lopez Ray Lopez
    3. February 2015 at 07:18

    OT- for your consideration, monetarists, a short FAQ of how money works. Once you understand how fragile the system is, you’ll hesitate to do radical surgery on it. In particular read the paragraph “How is *new* money created in a fractional reserve system?” This is my own work, not a screen scrape or Wikipedia cut-and-paste.

    How much U.S. currency is in circulation? There was approximately $1.34 trillion in circulation as of January 7, 2015. This is 1.34/16.8T = 8% of the total US GDP. As of the year 2005, the US currency in circulation was $375 billion (3.57x less than today, a 15.2%/yr increase), and more than half this amount was believed to reside outside the USA.
    What is the present US net worth? $80.7T (not sure if this includes public debt) What is the present US National Debt? $18.1T What is the present total US Debt?: $59.3T What is the current value of M2 money supply? $11.6T

    How is money expanded in the US fractional reserve system by the Fed? Three ways: Modifying reserve requirements, changing the discount rate, and open market operations. The total maximum money that can be lent by all banks, given an injection of X dollars in high-powered money into the Fed member banks, and a reserve requirement of r, is of course X*(1/(1-r)), so if r=10% you can get ten times the money loaned from any X dollars injected.

    How is *new* money created in a fractional reserve system? Three main ways (really just two, actually). BTW I don’t mean new money is ALWAYS created in these two main ways, since the Fed has ‘inventory’ of already existing money, just that these two ways are entry points for the creation of new money in an economy. First, the main way, is by the interest given by the US government when US government paper is sold to the public. The interest can be simply printed as ‘new money’. The mechanics would be: US Treasury sells paper to primary dealers, who then sell the paper to the Fed for dollars. Since the net dollars into the primary dealers from the government is net positive (due to interest), it’s new money. Second, another smaller way new money can be created by the Fed is when Fed member banks borrow money from the Fed at the discount window, called borrowed reserves, since Fed member banks must supply collateral. The collateral however can be not only unsecured government paper but also commercial paper such as an IOU from a consumer loan (secured by something tangible like a car or a house). This is also new money since this discount window borrowed reserves can increase over time, depending of course on public demand for money. Does anybody have stats on how these discount window borrowings have increased historically? I am interested in pre-2008 data, since post-2008 the Fed started buying huge amounts of mortgages which distorts the ‘new money’ data. Finally, a third way to create new money, and the least important, is simply to print money (cash, coin) and give it to banks, unsecured apparently, just to meet seasonal fluctuations; but I am told this ‘cash money’ is very small, and I’m not entirely clear on the mechanics of how this new money is distributed, but I suppose it is simply given unsecured to Fed member banks? Anybody know? This third way of creating new is apparently very small, trivial at the moment.

  20. Gravatar of TravisV TravisV
    3. February 2015 at 08:25

    Dear Commenters:

    Do Market Monetarists fully agree with this story?

    “The single currency itself caused the original Eurozone economic crisis”

  21. Gravatar of collin collin
    3. February 2015 at 08:49

    The ‘trils’ hits me as a solution that creates many more problems than it solves. Just like taxing mile driven versus collecting a gas tax which is much easier. Bond markets in developed markets work well enough. Most US government data is very good but there is a ton of noise in the data that anybody can poke holes in the information. We already have a hard enough time getting the market to accept inflation rates, unemployment and GDP number without bringing more money into it. (Have the Prices of TVs really gone down ~500% since ~1981 or did Prez Obama doctor the October 2012 jobs report for the election? In 10/12 there were a few people doubting the data but now 10/12 jobs report looks in-line with historical reality.) The idea of tying things to GDP hits me as lawsuits waiting to happen.

  22. Gravatar of ssumner ssumner
    3. February 2015 at 08:56

    Chacokevy, I don’t recall.

    SG, Good point.

    Nick, My understanding is that intermediate goods are included in GDP, unless then get put into final goods during the same year. If they go into inventories they are included. Perhaps you mean they could choose to include both durable consumer goods and the flow of services from durable consumer goods.

    Transfers certainly should not be included.

    Ray, Greece has its own monetary policy? There are things so stupid they are no longer even funny.

    Thanks Lars, I didn’t know that.

    Blue Eyes, Actually real estate does not track NGDP all that closely over the cycle, although perhaps it does in the long run.

  23. Gravatar of Charlie Jamieson Charlie Jamieson
    3. February 2015 at 08:58

    Ray, almost all money is created in the private banking system.
    When a bank lends you money, it doesn’t give you somebody else’s deposit, it creates a separate deposit for you.

  24. Gravatar of Jeff P Jeff P
    3. February 2015 at 08:59

    Scott —

    You know Varoufakis was Valve’s economist?

  25. Gravatar of ssumner ssumner
    3. February 2015 at 09:01

    Blue Eyes, Good point about Brown.

    Matt, Greece needs structural reform far more than anything else right now (assuming they stay on the euro.)

    Ray, You said:

    “As of the year 2005, the US currency in circulation was $375 billion”

    Please tell me you aren’t really this stupid. I don’t even need to look that up to know it’s wrong. Have you no common sense?

    Collin, We have TIPS already, and NGDP data is far harder to manipulate than the CPI.

  26. Gravatar of Nick Rowe Nick Rowe
    3. February 2015 at 10:25

    Suppose a firm buys a $40 software package from another firm, and uses that software to produce a final good worth $100.

    We could say that the software is an intermediate good, so value added and NGDP is only $60. Or we could say that the software is a final investment good, that will still be around next year, so NGDP is $140.

    NNetDP tries to get it straight by subtracting depreciation from NGrossDP. So it depends on whether the software depreciates or not. But measuring depreciation is always a judgement call, since it depends on expectations.

  27. Gravatar of Jesse Jesse
    3. February 2015 at 10:43

    Am I crazy to prefer a simple write-down? I think that NGDP-linked bonds are a pretty wonderful idea in general, but one the major desirable factors with NGDP-linked bonds is muted by the fact that Greece does not control its monetary policy. It almost feels like a “do right by me and I’ll do right by you” type of policy.

  28. Gravatar of Randomize Randomize
    3. February 2015 at 10:49

    Now if only the US would issue NGDP bonds and the Fed would target the spreads with regular bonds…

  29. Gravatar of Carl Carl
    3. February 2015 at 11:17

    For the record, there was $750 billion not $375 billion in circulation in 2005 per I was surprised by how steady the rate of growth has been for the period charted (1985-2015).

  30. Gravatar of Matt McOsker Matt McOsker
    3. February 2015 at 11:52

    Scott, not sure what structural reforms Greece can implement, or if anything will have the same impact as it would if they had their own central bank and currency. Greece is simply more like Detroit, without a lot of options.

  31. Gravatar of Morgan Warstler Morgan Warstler
    3. February 2015 at 12:07

    OK what does everyone think this is:

    “securities indexed to nominal economic growth”

    What would be carrot and stick for Greece with this setup?

  32. Gravatar of James in London James in London
    3. February 2015 at 12:39

    Re Varoufakis as an economist. I wouldn’t get too excited. He has published a book on the cause of the GFC, but here is an example of his reasoning:

    Something to do with US economic hegemony, I think. Enslaving the world by sharing its wealth, but not sharing it enough and hence collapsing. Or something.

    I think Simon Wren-Lewis would dismiss this as “heterdox economics”, ie cranky.

  33. Gravatar of ssumner ssumner
    3. February 2015 at 14:15

    Jeff, Yes.

    Nick, OK, I see your point.

    Jesse, What has the new Greek government done to deserve a write-down? If anything, the debt should be increased.

    Matt, How about free market capitalism? Greece is the least capitalist of all the developed countries.

    James, That doesn’t sound too promising.

  34. Gravatar of Major.Freedom Major.Freedom
    3. February 2015 at 14:40

    It is unfortunate that NGDPLT theory depends vitally on highly unreliable statistics, such as the “official” published rate of unemployment.

    I am hearing oblivious furvoring concerning the “official” rate of unemployment being 5.6%, in the financial press and elsewhere.

    I never see in these publications, which to a large degree are used by people who drive markets, the massive caveats in these stats. For example, if the job prospecting is so bad that job seekers give up and haven’t looked for work for just 4 weeks, then they are no longer counted as unemployed. Also, if you’re an out of work healthcare worker, retail manager, engineer or construction worker, if you work just one hour a week, for at least $20, then you are also not counted as unemployed. If you are a chemistry or math graduate who can’t find full time work but work 10 hours a week part time in a coffee shop, you are not counted.

    These issues have lead to false inferences of such things as the effectiveness of “stable” NGDP, since what causes stable NGDP causes unstable malinvestment and skill-capital mismatch, and the unemployment ignored above is not shown in the “official” stat thus making it seems like inflation is “working”.

    The past never repeats in economics. No comparison to the past is warranted.

  35. Gravatar of Derivs Derivs
    3. February 2015 at 15:49

    “The past never repeats in economics.”

    Whew! That’s a relief to know. Hyperinflation has only happened once and we never have to worry about it happening again.

    I can now go back to reading Mr. Varoufakis’ plan to accumulate experience points, level up, and gain powers that allow him to wipe away tranches of debt.

  36. Gravatar of Ray Lopez Ray Lopez
    3. February 2015 at 19:22

    Sumner: Ray, You said:”As of the year 2005, the US currency in circulation was $375 billion” Please tell me you aren’t really this stupid. I don’t even need to look that up to know it’s wrong. Have you no common sense?

    LOL! Are you trolling again? Look here: and go to Table 5, M1, “currency held by nonbank public” for 1995, it’s $366B. So I meant 1995, ten years earlier, which gives a 7.1%/yr increase. Thank you.

    I’m glad you agree with the rest of the post, with the implicit message that the fiat money system rests on a very narrow money base. If the Fed starts trying to create new money via the ‘second’ mechanism I described, borrowed reserves at the discount window, and starts buying commercial paper backed by kid’s lemonade stands, as the stylized hyperinflation example has it, then I think you’ll agree it will be the end of the world for us. Thank you.

    PS– (“Investors say GDP bonds won’t work”) – but you did get the knee-jerk consolidation prize of ‘nice-college try’: “This is an interesting academic exercise , but I believe it is unlikely that we will ever see widespread issuance of GDP-linked bonds,” said Mark Dowding, a senior portfolio manager at Bluebay, one of Europe’s largest bond funds. There you go. Market has spoke. Now you can go gently into the night, and drop this crazy idea, along with NGDPLT. Thank you.

  37. Gravatar of Prakash Prakash
    4. February 2015 at 01:44


    I wrote a comment on MR that basically answered the carrot and stick question. Cut off 25-35% of all government salaries and pensions and replace them with the same proportional value calculated (the same deal you’re offering the creditors), GDP based amount. If perquisites are a high percentage of the package, cut a higher percentage of the salary and less if otherwise.

    You don’t need to bother about all the intricacies of GDP calculation. All you know is that if the creditors hurt, the entire government and pensioners are hurting along with them.

  38. Gravatar of Jesse Jesse
    4. February 2015 at 07:10

    Scott, I don’t think writedowns are something that are earned or deserved. I think that the debt repayment is having a hugely negative impact on the Greek people. Would it have been worse absent the bailout? I’m not sure, but the underlying objective of the Eurozone is to create a union to promote peace through commerce. The bailout of Greece is a cost to that peace and I don’t think the people of Greece should be punished for poor government. Punish after the economy in Greece is back to full speed.

    My real point was more that I don’t love NGDP bonds for the Greek situation because they do not control their monetary system. Some might see that as a positive aspect of the NGDP bonds for Greece, but I think that the lack of control diminishes the value and will extend the focus on Greece and other periphery countries. With a writedown, there could at least be some finality to this mess. I think that the NGDP bonds create a conflict of interest in this particular scenario.

  39. Gravatar of Scott Sumner Scott Sumner
    4. February 2015 at 13:43

    Ray, When you make an idiotic claim that I point out is idiotic, there’s really no need for you to respond with a link telling us why you made the idiotic claim. We know why you make these claims.

    Jesse, The debt burden is significant, but certainly not the main problem in Greece. Greece needs to get out of the depression it’s in, which is not caused by the debt, but rather by bad economic policies leading to high unemployment. Greece needs economic reform.

    If the new government would offer the troika economic reforms, then the troika might want to lower the debt. But they are doing exactly the opposite, they are trying to roll back reforms made by the previous government.

    For democracy to work it is important that bad governance leads to bad outcomes and good governance leads to good outcomes. Whatever you choose to reward you’ll get more of. If you reward bad governance you’ll get more bad governance. If you reward good governance you’ll get more good governance.

    Unfortunately the Greeks have elected one corrupt government after another. I’m afraid that now they’ll have to pay the price.

  40. Gravatar of DOB DOB
    4. February 2015 at 13:59


    Noticed that Syriza’s open letter to German voters (which made a surprising amount of economic sense for something written by lefties) mentions nominal income/GDP three times?

  41. Gravatar of genauer genauer
    4. February 2015 at 14:33

    The result of Mr. Varoufakis meeting serious people at the ECB is the

    decision of the ECB to cut Greece immediately off the money tap

    just google WSJ “ECB suspends Greece” or words like that.
    (the mild greek version is “ECB fires warning shot by refusing to accept Greek gov’t bonds as collateral”

    I would expect the Greek bank runs to start tomorrow

  42. Gravatar of ssumner ssumner
    6. February 2015 at 08:53

    DOB, Yes, I saw that.

Leave a Reply