Inflation, jobs, and real GDP

I see that Paul Krugman is still insisting that Britain’s problems are due to fiscal “austerity,” which would only make sense if they were stuck in a liquidity trap and the BoE was no longer able to hit its nominal targets (whether inflation or NGDP.)

Even worse, he’s still using RGDP data even though NGDP is the appropriate way to ascertain the level of demand.  RGDP inappropriately mixes supply-side and demand-side issues. Employment in Britain is hitting record highs (unlike the US, where employment is mired far below the peak under Obama’s policies) suggesting that some of the low RGDP reflects supply-side problems.

Meanwhile, Mark Carney seems to have successfully talked the pound lower, and inflation expectations are soaring.

The U.K. 10-year break-even rate, a gauge of inflation expectations, climbed to the most since September 2008.

The rate, which measures the difference in yield between index-linked and nominal securities, rose six basis points to 3.35 percent at 12:43 p.m. London time.

For fiscal “austerity” to be a problem you’d somehow have to believe that it’s preventing the BoE from achieving even higher expected inflation rates.  How plausible does that sound?

PS.  Recall that famous example of liberal cluelessness at the New York Times?

Number in Prison Grows Despite Crime Reduction

The British newspaper The Guardian produced this gem:

Record jobs level with almost 30 million Britons in work

Unemployment down to 2.5 million by end of December but Labour says British workers paid price with wage cuts

In America employment is still more than 3 million below the pre-recession peak.  And we have a faster growing population than Britain.

PPS.  Nicolas Goetzmann sent me this FT article, which (implicitly) makes a strong case for the superiority of NGDPLT over inflation targeting, and then inexplicably ends as follows:

These considerations are highly relevant to the current discussion about the monetary policy framework that has been encouraged by Mark Carney, BoE governor-designate. To “tie the hands” of policy makers through a mechanism such as nominal GDP targeting could be unwise, as a shift in the relationship between inflation expectations and stock prices would require a change in policy.



42 Responses to “Inflation, jobs, and real GDP”

  1. Gravatar of Britmouse Britmouse
    12. March 2013 at 06:28

    Wadhwani’s FT piece followed up on a speech he made. a choice quotes…

    Mr Carney recognises some of the difficulties associated with a NGDP framework, not least the problems
    associated with determining the potential growth rate. If the nominal target does not change when the
    underlying real growth rate changes, for example, then this forces a change in inflation in the opposite
    direction. (Failure to acknowledge and take into account a slowing of potential GDP growth leads to a higher
    inflation target.)

    And that’s before he moves on to “helicopter drops”.

    This was nicer, from Stephen King:

  2. Gravatar of J J
    12. March 2013 at 06:46

    Slightly off-topic, but related to your post on Canada with the AD-AS diagram…

    Is the AS shock (from shrinking demand for Canadian exports) temporary? In the presence of a permanent AS shock, might it not make sense to slow NGDP growth? I understand that we don’t want to hit the economy with a AD shock as well as a AS shock. But, suppose half the US vanishes through a black-hole. Full-employment RGDP would (approximately) cut in half. Keeping NGDP growing at 5% per year would lead to incredibly high amounts of inflation. I understand that inflation is OK if there is a temporary AS shock (resulting from decreased demand from Americans because of a recession) because we want to keep NGDP stable to promote recovery. But, if there is a permanent real fall in what we can produce, then not reducing NGDP will just lead to higher inflation and maybe slow the process of output falling to a stable level.

  3. Gravatar of Inflation, jobs, and real GDP | Fifth Estate Inflation, jobs, and real GDP | Fifth Estate
    12. March 2013 at 06:48

    […] See full story on […]

  4. Gravatar of marris marris
    12. March 2013 at 09:05

    Sorry, also slightly off topic. What is the best write-up of how an NGDP futures market would work? I’m guessing that people take bets on the *actual NGDP level* at some future time T. So the CB target isn’t in the payoff calculation. For example, suppose that Ptarget is the target price for T (announced by the central bank) and Pforward is the current future price with fixing date T, and Pactual is the actual price (actual NGDP level) at time T. If I buy one share/contract/quantity now, then my long position makes (Pactual-Pforward). is that correct? And the central bank’s asset purchases try to keep Pforward around Ptarget?

  5. Gravatar of Tyler Joyner Tyler Joyner
    12. March 2013 at 09:11


    Your explanation is a little confusing, but I believe Scott’s proposed system looks something like that.

    Personally I think that a RGDP futures market makes more sense, one in which speculators and hedgers simply bet on future RGDP. Then the government stays completely out of the operation of the futures market, but rather uses the price of the contracts to “target the forecast” and adjust inflation to achieve the NGDP target.

  6. Gravatar of TallDave TallDave
    12. March 2013 at 09:18

    not least the problems associated with determining the potential growth rate. If the nominal target does not change when the underlying real growth rate changes, for example, then this forces a change in inflation in the opposite direction.

    That’s sort of the point, though — you want looser monetary policy when real growth is slow, and tighter money when it’s high. More importantly, you want markets to expect that. The lesson of the “liquidity traps” is that keeping NGDP expectations stable is more important than keeping inflation expectations stable (assuming you could even know what inflation was).

  7. Gravatar of Mike Sax Mike Sax
    12. March 2013 at 09:21

    Britain employment is at record highs? What’s the unemployment rate? I don’t think it’s any lower than the U.S.

  8. Gravatar of TallDave TallDave
    12. March 2013 at 09:27

    marris — the wiki isn’t bad, has several alternatives with cites.

  9. Gravatar of Tyler Joyner Tyler Joyner
    12. March 2013 at 09:27


    Wadhwani’s comment was talking about how to arrive at that 4.5% NGDP target, and how to know if and when the trajectory of real growth changes. I don’t think most African nations would be well served by a 4.5% NGDP target, as an easy example, so how does one decide what the target should be? Throw darts?

  10. Gravatar of TravisV TravisV
    12. March 2013 at 09:35

    Prof. Sumner,

    StatsGuy spelled out an argument for why the stimulus program’s government spending did increase NGDP in 2009 (not sure whether he agrees with it). Here is what he wrote:

    “The Fed is politically constrained. This was the argument in 2009ish. That is, the Fed could not engage in the magnitude of action required due to political reasons, but it did have the political lattitude to accommodate an increase in fiscal expense. Hence the “Fed can’t carry this burden alone” argument that Bernanke at times put forward.”

    Do you agree with this argument? Did the stimulus program’s government spending increase NGDP in 2009?

  11. Gravatar of TallDave TallDave
    12. March 2013 at 09:37


    Most African nations are kleptocracies of one flavor or another (Botswana being a notable exception) with much bigger problems than the difference between inflation/NGDP targeting.

    That said, I don’t think it’s terribly difficult to look at real growth trends and derive a relatively simple formula based on expected higher catch-up growth rates in developing economies.

  12. Gravatar of Tyler Joyner Tyler Joyner
    12. March 2013 at 09:40


    About the wiki: Yeah that confirms what I thought were the existing ideas. To me they all suffer from a lack of understanding of futures markets.

    From the wiki,

    Idea #1 “Investors would initiate trades as long as they saw profit opportunities from NGDP growth above (or below) the target. The money supply and interest rates would adjust to the point where markets expected NGDP to reach the target”.

    So investors are basically going to take positions based on their NGDP predictions, at which point the Fed will specifically intervene to bring the forecast back to the target. Who the heck would take a position when they know the Fed is on the sidelines with a limitless supply of money ready to move the price against you?

    Idea #2 “Bill Woolsey offers several alternatives for the structure of such a futures market, suggesting an approach in which the Fed maintains a fixed price for the futures contract, hedging any resulting short or long position by conducting OMOs to match its net position and using other traditional techniques such as changing reserve requirements.”

    So now we expect people to take positions in an asset with a fixed price. Why would they do that?

  13. Gravatar of TallDave TallDave
    12. March 2013 at 09:45

    1. Because the Fed could be wrong. Omnipotence isn’t omniscience.

    2. I think he means the Fed’s price is fixed. If you think the Fed is wrong, you buy/sell.

  14. Gravatar of TallDave TallDave
    12. March 2013 at 09:53

    I mean, the Fed targets inflation today, but people do bet against their targets. Look at what Giustra’s doing.

  15. Gravatar of Tyler Joyner Tyler Joyner
    12. March 2013 at 09:55

    1. The Fed could be temporarily wrong, but they explicitly say that if they are wrong, they will step in and move the markets back to the target. So the trader is expected to first take a bet that the Fed will fail to meet the target and then also get out of that position before the Fed intervenes against him. You’d have a better chance of making money speculating at a slot machine.

    2. If the Fed is offering a fixed price and has no limits to the size of the position it is willing to take, that will be the price of the contract. There is no profit potential.

  16. Gravatar of Tyler Joyner Tyler Joyner
    12. March 2013 at 10:01

    “I mean, the Fed targets inflation today, but people do bet against their targets. Look at what Giustra’s doing.”

    So NGDPLT is being sold in large part due to the benefit of having a predictable monetary policy, but the mechanism upon which it rests relies on traders and investors believing that the Fed will fail to achieve the target? And the more the level targeting works, the more consistently the traders who provide liquidity to the futures market will lose money.

  17. Gravatar of TallDave TallDave
    12. March 2013 at 10:02

    1. Yes. Again, people already bet against Fed targets all the time. Yes, even though the Fed can intervene.

    2. hedging any resulting short or long position

  18. Gravatar of marris marris
    12. March 2013 at 10:09

    @TallDave Thanks for the link. I read that page already, but didn’t check out the references. I will take a look. I assumed there was some MM FAQ somewhere, but I couldn’t find one.

    BTW, I share Tyler’s confusion about the NGDP futures market. The central bank (CB) offering to buy or sell contracts at $X seems very similar to the gold standard, where the central bank bought and sold gold for $X per ounce. Unless there’s a run on the central bank, the price will stay at $X per ounce. In the case of NGDP, the Fed can make contracts at will, so I won’t see why there would be a “run” of any kind… so the future price will stay fixed at $X.

    Put it another way. Suppose I thought actual NGDP was going to be $Y where Y>X. How could I profit from this? Would I buy contracts, expecting to make $(Y-X) on them? Would the Fed take my money and buy assets with them? Would they sell assets?

  19. Gravatar of Tyler Joyner Tyler Joyner
    12. March 2013 at 10:15

    1. This is not comparable, and you should know it. People are willing to take bets on inflation for a couple of reasons, not the least of which is because the Fed changes its inflation target. The Fed also has multiple mandates. So the speculator can make money if he anticipates the Fed will change its inflation target, or if he believes that the Fed will prioritize employment over inflation.

    NGDPLT proposes to maintain one single target and have the Fed pursue only one mandate, and its proponents believe NGDPLT is superior precisely because it will succeed at hitting that target. So either it doesn’t work and it’s useless, or it works and it puts the speculators out of business in short order, at which point it doesn’t work.

    2. Correct. Meaning the Fed will take the opposite side of every trade, or take the opposite side of a net long or short position of the non-Fed market participants. Meaning the price will never change. Meaning it’s impossible to make any money by speculating on it.

  20. Gravatar of TallDave TallDave
    12. March 2013 at 10:28

    1. It’s not that different. One could (for instance) bet that the Fed will change targets, or abandon NGDPLT altogether.

    I don’t think anyone expect the Fed to be so successful that NGDP will always be exactly what they target, nor is it sensible to conclude there’s no room between omniscience and failure.

    2. Presumably the hedges and the contracts would have different counterparties. One side would make money, the other would lose it.

    Scott or Brad probably has a better post on the mechanics, I should try to find it when I have more time.

  21. Gravatar of TallDave TallDave
    12. March 2013 at 10:38

    OK, that was easier than I thought.

    To give a simplified overview, the Fed would create NGDP futures contracts and peg them at a price that would rise at 5% per year. If investors expected NGDP growth above 5%, they would buy these contracts from the Fed. This would be an “open market sale,” which would automatically tighten the money supply and raise interest rates. The Fed’s role would be passive, merely offering to buy or sell the contracts at the specified target price, and settling the contracts a year later. Market participants would buy and sell these contracts until they no longer saw profit opportunities, i.e., until the money supply and interest rates adjusted to the point where NGDP was expected by the market to grow at the target rate.

    Interestingly Scott alst makes the gold standard comparison:

    It might be helpful to compare this idea to the old international gold standard. Under that system, the U.S. government agreed to buy and sell unlimited gold at $20.67 per ounce. This kept gold prices stable, and the money supply adjusted automatically. Unfortunately, however, stable gold prices did not always mean a stable macroeconomic environment. Putting NGDP futures contracts on the market along a similar model would likewise create a stable price for those contracts, hence stabilizing expected NGDP growth. And stable NGDP growth would be more conducive to macroeconomic stability than a stable price of gold, especially in a world in which rapidly growing demand from Asia might distort the relative price of gold.

  22. Gravatar of Scott Sumner Scott Sumner
    12. March 2013 at 10:38

    Thanks for the link Britmouse. That’s a common mistake.

    J, It’s not a problem if lower RGDP leads to higher inflation. However I do think we should target NGDP/person, so that addresses the case where half the population disappears. But if RGDP falls due tolower productivity, then higher inflation is actually desirable.

    marris, That’s right if I understand you correctly. I have a paper coming out soon on NGDP futures.

    Tyler, African nations have a higher population growth rate, so 6% would be better.

    TravisV, I think it increased NGDP in 2009, but probably not 2010, or thereafter.
    Might have even reduced it in 2010.

  23. Gravatar of TallDave TallDave
    12. March 2013 at 10:50

    I have a paper coming out soon on NGDP futures.

    Well, that should help. I admit the NGDP futures contracts are a bit confusing to me too — do the profits partially result from the monetary change itself (i.e., inflation/deflation)?

  24. Gravatar of Tyler Joyner Tyler Joyner
    12. March 2013 at 10:51

    1. Again, what we’ve been told is the point of NGDPLT is that the target will not change, but rather than NGDP growth will be on a steady path. Just because it’s hypothetically possible to make money under some circumstance does not mean you have the basis for a viable, on-going market here. Why participate in a market when your chance for making money specifically relies on the market failing to serve its intended purpose? If they abandon NGDPLT, then at the very least the traders and firms who specialize in that market will be out of business. The entire concept is perverse.

    2. Again, idea #2 specifically states that the contract will not change in price, but rather that the Fed will have different long or short positions based on the other market participants. So while yes, it’s probable that some trades will happen between two speculators, as a whole the market price of the contract will not change because the Fed will specifically and intentionally counteract any movement in the price, and has unlimited power to do so.

    And this really does not need to be a stumbling block for NGDPLT. What a futures market requires to function is a reasonable chance for speculators to make money. This can be accomplished by using a RGDP futures market. The traders may know that the Fed wants to hit a NGDP target and will adjust inflation to do so, but the portion of that NGDP target which is RGDP is undetermined. That is a worthwhile object of speculation, and even one for which there may be significant existing demand which is not being met. Let’s consider the Fed’s angle and the speculator’s angle.

    Fed: If the projected RGDP increases, as determined by an increase in the futures contract price, then the Fed will need to decrease inflation to maintain the NGDP target. If projected RGDP decreases, then the Fed must increase inflation. Movements in inflation will of course affect RGDP to some degree (if it didn’t then the whole point would be moot), but RGDP moves due to a large number of factors, of which inflation is only one.

    Speculator/hedger: If the current projection of RGDP is higher than what I think RGDP will be, there is a potential for profit. I know that the Fed may intervene in terms of inflation, but the effect on RGDP in the short term will be minor, and I can calculate how that might affect my position.

    If I’m a cyclical business, then it may benefit me to hedge my risk by taking a short position in RGDP futures, because I know that if RGDP moves down then my business will suffer.

    If I’m a business that suffers unduly from inflation, I might be inclined to take a short position in RGDP futures, because I know that if RGDP goes down, the Fed will increase inflation, so the loss in my real business is offset by a gain in my futures position.

    Et cetera.

  25. Gravatar of TallDave TallDave
    12. March 2013 at 11:19

    Tyler, I think what you might be missing is that the contracts themselves create the monetary effect — the Fed does not otherwise intervene (it doesn’t have to). If there’s no profit to be made, then the Fed has achieved the policy. If there’s profit to be made, then the Fed is acting in support of the policy by offering the contract.

    As long as the Fed isn’t omniscient, there should be a chance to make money.

    I guess I’ll wait for Scott’s paper before confusing myself any further 🙂

  26. Gravatar of StatsGuy StatsGuy
    12. March 2013 at 11:31

    Travis –

    To be clear, this implies you believe that there is a multiplier effect (a la Woodford). ssumner’s argument is that even if there IS a multiplier effect, the Fed has the last word, so if the Fed commits to a specific target, all the fiscal authority does is shift AD from the private to the public sector (with a deadweight loss)

    IF there is a true multiplier effect (and I’m not arguing for/against), AND the Fed has an asymmetric response function, THEN the fiscal authority can increase AD even if it’s no more efficient at the margin than private sector activity.

    There’s other odd cases too… Krugman’s argument that even if the Fed wants higher NGDP and declares a target it won’t be credible, but that a fiscal expenditure is always credible, for example (if you believe it). Or, for example, if there are no assets to invest in (although we long joked there’s no risk of that, it’s noteworthy that the Fed MIGHT actually run out of long term safe assets to buy with enough QE, but hopefully that will be because a rising economy increases tax revenues).

    Don’t let these odd cases hide the core of ssumner’s argument. He’s basically 98% right.

  27. Gravatar of StatsGuy StatsGuy
    12. March 2013 at 11:33

    I think the question for Scott is:

    “Let’s assume the Fed really was politically constrained in 2009, and really could not do more – IF TRUE, were we better off with the stimulus than we would have been without it?”

    Things were pretty dark in 2009… I don’t know the answer to that. Among other things, it requires understanding the belief response function (psychological impact of the stimulus failure to pass). Scott may have an opinion.

  28. Gravatar of Tyler Joyner Tyler Joyner
    12. March 2013 at 12:12


    I’ve asked Scott for more specifics on how the market works before, but haven’t gotten much more than what’s already in the wiki you linked.

    This is just my hypothesis, but I’m guessing that Scott, Bill, et al might not have a ton of experience in the actual functioning of the futures markets we have today. Creating a NGDP futures market doesn’t necessarily mean it has to look exactly like the ones that already exist for other products, but they were designed the way they are for a reason, and those reasons at least should be considered.

    So let’s consider the possibility you just raised – that the NGDP futures market is designed to work via speculators putting money into it (going long) or taking money out (going short), which automatically expands and contracts the monetary base. In that scenario the Fed is acting as counter party to the transactions and the price of the contract wouldn’t change – only the open interest (the number of outstanding contracts). That “works” in the sense that it is a way to automate the expansion/contraction of the monetary base. The problem is that there is no motivation on the part of the speculator/hedger to take part. If the price doesn’t change, there is no profit. If there is no profit, there is no speculators, and if there is no speculators, the market loses liquidity and ceases to function properly.

    Another significant problem with that process is margin. Regardless of whether I’m short or long, I’ve got to put margin down to take a position in the futures market. The only money that actually gets added to or taken from the monetary base is in the form of profits or losses made on the trade. It’s not like you go short a NGDP contract with zero money down and the Fed hands you a million bucks to go add to the monetary base. The money stays put until you close out your position, at which point you get what you put in plus or minus gains/losses.

    So, basically, the contracts themselves are not going to change the monetary base. For every futures contract there is a long and a short – even if the Fed is creating new money to back up their side of the trade, that money vanishes when the speculator exits the position. The only “new” money would be profits on the trade.

  29. Gravatar of ChargerCarl ChargerCarl
    12. March 2013 at 12:51

  30. Gravatar of Frances Coppola Frances Coppola
    12. March 2013 at 16:10

    Mike Sax,

    UK does indeed have unemployment levels of about the same as the US. But we have quite a lot of people entering the workforce, especially over-60s coming out of retirement because of squeeze on fixed incomes, and people who have previously been on long-term disability benefit. We also have people taking second jobs, and employment figures include young people doing unpaid internships and people on Workfare (we don’t call it that, but that’s what it is). All of which adds up to an interesting combination of highish unemployment AND record employment.

    In my view we have disguised unemployment. Older people and people with disabilities coming into the workforce find it very difficult to obtain employment, but often can find casual and occasional work, so they register as self-employed – there is anecdotal evidence that this is encouraged by Work Programme staff. Consequently we have a large rise in self-employment even though self-employed incomes are crashing: ONS suggests the fall in self-employed incomes is at least partly caused by the influx of new entrants in a stagnant market. There is a particularly large rise in self-employment among women wanting to work part-time in the interim management market, which suggests to me that many of the people leaving public sector employment and financial services are obtaining occasional consultancy work. Proportionately more women than men have lost their jobs in both the public sector and financial services.

    We also have poor productivity and a falling median income, disguised by rising incomes among the top earners. Mean income now exceeds median by quite a bit. As in the US, semi-routine mid-skill jobs are disappearing due to automation, while there are shortages in high-skill jobs, and low-skill jobs become increasingly short-term, part-time and insecure. Exactly why productivity is poor is something of a puzzle, but it does seem to relate to the falling income level. So apparently healthy employment disguises less healthy indicators.

    Scott, sorry, I’ve hijacked your blog! Do you have any views on the UK’s productivity puzzle and the interesting combination of high unemployment and record employment?

  31. Gravatar of Rademaker Rademaker
    12. March 2013 at 17:11

    “Employment in Britain is hitting record highs (unlike the US, where employment is mired far below the peak under Obama’s policies) suggesting that some of the low RGDP reflects supply-side problems.”

    A taste of things to come for the united states maybe? Is all this QE really boosting demand or just generating inflation of the cost-push kind by handing out capital to bond market front-running off-shore entities that express wealth-effect generated demand through other currencies?

  32. Gravatar of TallDave TallDave
    12. March 2013 at 20:24


    Well, we’ll both look forward to Scott’s paper then 🙂

    I think there should always be opportunity for profit in the spread between 5% and what the Fed actually achieves, and buying the contracts moves money into the Fed, while selling them moves money out — margin doesn’t get deposited with the Fed.

  33. Gravatar of W. Peden W. Peden
    13. March 2013 at 02:02

    Frances Coppola,

    Much (but not all) of the story concerns factors affecting North Sea Oil revenues-

  34. Gravatar of ssumner ssumner
    13. March 2013 at 04:23

    Statsguy. I define “doing more” as allowing more NGDP growth. So if it was constrained and could not allow more NGDP growth, then fiscal stimulus would have had zero effect.

    As an aside, the Fed was not politically constrained in 2009, they were internally constrained. Everyone could see in early 2009 that there was a risk of depression. The stock market had crashed. If the Fed had announced something like 2% inflation targeting, level targeting, no one would be accusing them of hyperinflation. But of course that would have been far more expansionary than the policy they actually did. That policy would have MORE THAN offset the complete lack of any fiscal stimulus.

    Frances, I’ve discussed that elsewhere:

    1. Falling oil output.

    2. Falling finance output

    3. Big government (which grew sharply after 2000)

    Rademaker, I believe that monetary stimulus boosted output in the UK, those extra workers are presumably producing something.

  35. Gravatar of Tom Tom
    13. March 2013 at 04:58

    As you continue the good fight to target NGDP, let’s also push for better measures of “tight” and “loose” money.
    When money is tight, small business loan applicants get rejected; they get loans when money is loose.

    The SBA explains how banks have kept money tight:
    “It is strange to note that, over 50% of Small Business Loans applied by the entrepreneurs’ to the Banks are rejected for some reasons, where as the Federal government has done it job to help the small business enterprises. More than 30 billion dollars have been sanctioned to banks to provide loans to small business as and when they need it and apply for it.The banks have not been helpful, instead they utilized the money to clear there own TARP. Banks used over 26 billion dollars to clear their obligations and only about 4 billion dollars were actually given as small business loans to the entrepreneurs. A survey conducted by Pepperdine University to find out the exact amount given as loan to small business enterprises, revealed that more than 60 percent application for loans had been turned down, which is a very high rate of rejections.”

    Market Monetarists should be tracking the SBA loan rejection rate to discuss how tight or loose money is.

  36. Gravatar of Tyler Joyner Tyler Joyner
    13. March 2013 at 06:39


    Yes, I’m interested to read the paper.

    I think we’re starting to go in circles at this point, but for clarity’s sake I’ll give it one more go.

    This: “I think there should always be opportunity for profit in the spread between 5% and what the Fed actually achieves”

    is wrong. It violates the most basic understanding of how the financial markets work. Betting on what you think the Fed will do is totally different from sitting down at the poker table across from the Fed and trying to beat it. You do not win against someone with a limitless supply of money and the spoken willingness to use as much as necessary to insure that you lose.

    Envision this scenario, and understand that this is what you’re describing:

    Trader A knows that the target NGDP is 5.0%, and because the Fed intervened yesterday the current contract value is at the 5.0% level. When the Fed intervened yesterday, A’s friend Trader B lost all of his money and all of his client’s money, because the Fed came in like a nine hundred pound gorilla and pushed the market against Trader B’s positions. Trader A can go either long or short from here, but he knows that in either case the Fed will come in and move the needle back to 5.0%. The only remote hope for a profit would be to take a position, make a little money, and then get out of it before the gorilla arrives.

    Or, Trader A could go back to trading wheat futures, in which the government never intervenes, where there are co-ops and grain elevators who need to lay off their risk, and where Trader A previously made a good living for 20 years.

    I can almost guarantee that Scott’s futures paper describes something different from what the wiki says, because anyone who has ever traded futures or been involved with the industry would take one look at what’s on the wiki and say exactly the same thing I’ve said.

    That said, this:

    “buying the contracts moves money into the Fed, while selling them moves money out “” margin doesn’t get deposited with the Fed”

    is also incorrect. You don’t buy a contract – you go long a contract. You deposit margin with the exchange, and your gains or losses are marked to margin. You can also go short a contract, in which case you deposit exactly the same amount of margin as if you went long. The contract is nothing more or less than an agreement to deliver a specified good on a specified date in a specified quantity, and it must be of a specified quality. In the case of financial futures, as a NGDP contract would be, you agree to settle in cash the difference between the price when you take on your position and the price on the delivery date.

    So, there are really only two possibilities here. One is that the Fed is using the futures market as a price discovery mechanism only, and actually doing the OMOs and other operations in the manner we’re used to. The other possibility is that the only money that actually goes into or comes out of the base as a result of the futures market is in the form of gains and losses on the trades. There is no mechanism in a futures contract itself which would change the monetary base.

  37. Gravatar of Tyler Joyner Tyler Joyner
    13. March 2013 at 07:12

    I think if you take the Swiss currency intervention as a similar situation you can see this in action. They told the markets and the world “our currency will not appreciate beyond 1.2 Euros to the franc. We will purchase unlimited amounts of Euros to ensure this”.

    Since then, the franc has never gone beyond that ceiling. It comes close, has dipped fractionally below that intraday on one occasion, but basically the Swiss did what they set out to do. Banks and other participants in the FX market are not willing to get crushed by the SNB, so once the exchange rate gets to 1.2, the pool of buyers of the CHF/EUR evaporates.

    Now understand that what the NGDP futures market apparently proposes to have a ceiling and a floor which are both maintained with the same strength and commitment that the SNB showed toward its exchange rate. And the ceiling and the floor are touching each other. The result will be zero liquidity, because no one would participate in such a rigged market.

    As I said before, I’m fairly sure Scott’s paper will be something different than this.

  38. Gravatar of Tyler Joyner Tyler Joyner
    13. March 2013 at 07:12

    *1.2 francs to the Euro, rather.

  39. Gravatar of Doug M Doug M
    13. March 2013 at 08:30

    Tyler, Tall Dave,

    After reading the paper, I wouldn’t call it a futures market. What the professor describes is more akin to a warrant. The fed will be selling a security that will pay the difference between NGDP (realized) and the NGDP target.

    And the Fed should let players make a naked short on these warrents, without posting additional collateral or requiring the proceeds be held at the Fed. This could create some massive counterparty risk.

    Argentinia issued some GDP linked warrents — the Argentine warrants distribute a fixed payment to investors every year that Argentia RGDPLT stays above 3%. These were issued to Argentina’s creditors as they exited the 2001 bankruptcy.

  40. Gravatar of Tyler Joyner Tyler Joyner
    13. March 2013 at 08:42

    Doug M,

    Do you have a link to the paper? In looking for it I found a 2009 post by Scott that goes into much more detail on the futures market. I’m reading through that now.

  41. Gravatar of Tyler Joyner Tyler Joyner
    13. March 2013 at 10:11

    The 2009 post has a lot more meat to chew on than the posts we’ve seen since I started following this blog about 6 months ago. I’m not sure why it didn’t turn up in my previous google searches on the subject, but whatever.

    To address the points TallDave and I were debating, and this is assuming the 2009 post “Spot the flaw in nominal futures index targeting” is more or less the same policy regime Scott advocates now…

    Looks like the futures contract itself is not the mechanism for changing the monetary base, but rather a price discovery mechanism which the Fed uses to decide which OMOs to undertake, as I thought.

    He does indeed say the price of the contract will be set and the Fed will buy or sell them in unlimited quantities according to demand, so that’s answers that question. I think, however, that the more appropriate comparison would be to a strike price on an option. The traders are really only saying whether they think NGDP 1 year from today will be above or below today’s target.

    There is no reason that I can see why the actual price the traders pay would have to be a direct ratio of the NGDP target. You could just as easily specify a strike price and then set the margin requirements at whatever you want. $100 would make sense, as Scott specified a $500 OMO for every $100 put into the futures market.

    Although Scott specified that every day brings a new contract, one thing he wasn’t clear on was whether previous contracts were still available to be traded all the way up to the delivery date – my impression was that they would not be. This is both good and bad.

    It’s good because having 14 months of prior contracts available to trade every day would have serious implications on the liquidity of the market. Even in very liquid futures markets for things like the S&P 500 E-mini, there is only one contract for each quarter, and at any given time only two of those are being actively traded.

    It’s bad because the utility of buying an investment that you cannot get out of for 14 months is pretty limited. At least with normal futures you can choose between the front month and months farther back. The way Scott’s 2009 post reads, if you want to bet on what NGDP will be on January 4th 2015, you’ve got to buy a contract on January 4th 2014 or not at all. Combine that with about 250 contracts per year (250 trading days most years), and you have serious liquidity issues.

    To partially counteract this, Scott proposes to pay interest on margin accounts. A few issues come immediately to mind with that. First, to overcome the difficulties I outlined above the interest paid would have to be relatively high. And yet if the interest paid is too high, the investor no longer cares as much about getting the NGDP guess right, but is more likely to put on equal long/short positions (spaced a day apart) to lock in the interest payments without exposing oneself to any directional risk.

    Last but not least, the points I made about betting against the Fed still apply, if to a lesser extent. Since the contracts price is fixed and the payout contingent upon NGDP a year in the future, the scenario I laid out where the Fed pushes the price of a contract and crushes the traders is moot. That being said, the validity of the market is still based upon a paradoxical relationship between the confidence and stability NGDPLT is supposedly conferring upon us, and incentives for traders to take part in this market resting solely on the possibility that the Fed will fail in its stated goal of maintaining NGDP growth at a specific target.

    Doug M is right though, it’s not really a futures contract. It’s not an option or a warrant either though. The scheme by which profits are determined is similar to a European style option, but even those can be traded freely up until the expiration date.

  42. Gravatar of ssumner ssumner
    14. March 2013 at 12:08

    Tyler, Yes, the “spot the flaw” post is a good intro.

Leave a Reply