On the irrelevance of near-monies
Tyler Cowen links to a very good post by Cardiff Garcia:
It’s no longer enough to understand traditional monetary policy transmission mechanisms (money multipliers, federal funds rates, reserves); it is also necessary to understand how the shadow banking system (collateral supply, rehypothecation) affects monetary policy, and vice versa.
Mr. Garcia is correct that the shadow banking system would be important if money multipliers, fed funds rates, and reserves were important. But they aren’t. All that matters (for AD) is the monetary base and base velocity.
Astute readers will argue that a big crisis in the shadow banking system could have a major impact on base velocity. There are three answers to that argument:
1. It’s not true in countries with high NGDP growth (or inflation) targets.
2. It’s probably not true in countries with 5% NGDP growth targets, level targeting.
3. Even if it were true, it’s not important.
The monetary base used to be called “high-powered money.” This was because it didn’t pay interest, while other safe assets like T-bills did pay interest. These two facts made base money a very undesirable asset for investors (except tax evaders.) As a result, an increase in the base would tend to raise NGDP roughly proportionately. If a country has a high NGDP growth target, it’s nominal interest rate will never fall to zero, the base will be more than 90% currency, and shadow banking crises will never significantly impact base velocity. I also believe that to be the case when you have 5% NGDP growth rate targeting, as long as it is level targeting, but I can’t be sure.
More importantly, the central bank has almost unlimited ability to boost the monetary base and offset a fall in velocity. There are risks of this sort of policy, but they are tiny compared to the risks incurred by other government policies such as TARP. People who worry about the zero bound are missing the bigger picture. A zero bound is a policy decision. There’s no point in crying about an economy at the zero bound. If we are there, that means our NGDP growth target is too low to prevent us from being there. Reality is reality, there is nothing we can do about it. We have to deal with it. If you don’t like the zero bound then raise the NGDP (or inflation) target. Some economists sound like crybabies—complaining that the zero bound is a big problem and also complaining that we can’t raise the inflation target because the central bank would lose credibility and all sorts of bad things would happen. Fine, but then your two choices are either printing lots of money or suffering a deep recession and debt crisis. Take your pick.
PS. This post isn’t really a critique of Cardiff, as he’s arguing that near monies matter if you are targeting M2. He’s right.
PPS. Perhaps before I arrogantly asserted that “rehypothecation” doesn’t matter, I should have looked up the word to find out what it means. But then I’m a hedgehog and Tyler’s a fox.
Tags:
6. April 2012 at 07:05
It also matters if the you are targeting some short and safe interest rate, and especially if you are targeting it according to a fixed rule to provide for creditibility.
For example, the federal funds rate should be equal to 1.5 times the inflation rate plus .5 times the output gap.
If you believe that changing 1.5 or .5 in the equation will be a violation of the rule, and this will destroy credibility, then perhaps the shadow banking system is a big worry.
It is also true, of course, if you are targeting the level of base money, M1 or M2. The “near monies,” which in my opinion are sometimes “money,” created by the shadow banking system are very important.
Of course, to the degree a target growth path for nominal GDP fixes the problem, well, that is just one more advantage of the target rule. (Still, I doubt that some rule like the federal funds rate equals .5 times the nominal GDP gap will be robust against changes in shadow banking, and so if someone things that we need such a rule to make the nominal GDP level target credible, shadow banking is still a problem.)
And to return to my fixation, along with a nominal GDP level target, private currency is the best solution to the zero negative bound. Short of that, a really big quantity of base money (including long term and maybe even risky asset on the central bank’s balance sheet,) and lowest on the list is a higher trend growth rate of nominal GDP (and so inflation.)
6. April 2012 at 07:47
Scott,
You said;
“All that matters (for AD) is the monetary base and base velocity.”
Yes, as in indicator. If people are spending more (consumption and saving/investment), they will need more money so the monetary base goes up.
What causes them to spend more is an increase in the price of financial assets relative to real goods/assets and expectations of higher NGDP growth.
6. April 2012 at 07:49
Is the conclusion that near-monies don’t matter a theoretical or an empirical observation? Or both?
You’ve said in the past “in fact the banking system doesn’t actually play an important role in monetary policy. It’s all about currency in circulation.” (That was here: http://www.themoneyillusion.com/?p=13411)
So in fact it’s not just near-monies that are irrelevant, but demand deposit obligations don’t matter either. Right?
This is where I stumble in coming around to your worldview …
6. April 2012 at 07:53
Tyler thought we would be hearing a lot about rehypothecation in 2012 and so far we haven’t, but then the year is still young. Expectations matter and a couple of factors make this scenario a bit iffy. Garcia stressed how (the new paradigm of)asset markets have redefined the role of banking institutions. However, it’s too bad those asset markets were ever called ‘shadow banking’ in the first place, especially as the role of these markets becomes increasingly important. Also, it does make some people nervous that – as Mehrling said, the Fed came to be not just the lender of last resort in the crisis, but the dealer of last resort. Perhaps the fact that the Fed now holds cards for every player in the game is key. Of course while that is scary for some, it does make it a lot harder to take the game away from the Fed. I never said I was foxy. But I am a fox.
6. April 2012 at 07:59
Scott
To me K Smith is talking in an unknown foreign language about the same Cardiff Garcia piece:
http://modeledbehavior.com/2012/04/06/neo-wicksellian-shadows/
6. April 2012 at 08:08
Marcus I noticed that too. It made we wonder whether there would be an argument for government becoming a larger component of the economy, coming out of all this.
6. April 2012 at 08:10
Scott, if cheques were used as medium of exchange as they used to be or if banks circulated their own notes as they used to do, I would say you’re wrong for sure.
But other than bankers acceptances and cheques in international trade, this era is basically over.
The goal in broad money aggregates was to identify an index with stable velocity. Is it your goal to dispute that or are you simply stating that casuality always ran from base -> ngdp -> broad money
Which makes broad money an indicator but a collapse in broad money due to financial panic irrelevant?
6. April 2012 at 08:19
Bill, I agree.
dtoh, I see the hot potato effect as the transmission mechanism that causes spending to rise. But asset prices can certainly affect V.
MR, Yes, I also don’t put much weight on demand deposits. The key is base money and base velocity. That’s based on both theoretical and empirical considerations.
Becky, George Selgin has a good paper on the dealer of last resort issue. I’ll do a post soon.
marcus, I’ll take a look.
Jon,. You said;
“Scott, if cheques were used as medium of exchange as they used to be or if banks circulated their own notes as they used to do, I would say you’re wrong for sure.”
No, I’d still say it’s base money that matters, because that’s what the Fed controls.
Yes, cause goes from base to NGDP to broader aggregates. I am not interested in finding an aggregate with stable velocity, because I don’t favor intermediate targets (other than NGDP futures.)
6. April 2012 at 08:27
I’m with Marcus. Maybe that crack from Karl about the view from 30K feet means he’s flying first class and the flight attendants are being too free with the complimentary wine.
Whatever. I think this paper must be what the rehypothecation argument is about;
http://www.imf.org/external/pubs/ft/wp/2010/wp10172.pdf
——-quote——-
Rehypothecation occurs when the collateral posted by a prime brokerage client (e.g., hedge fund) to its prime broker is used as collateral also by the prime broker for its own purposes.
Every Customer Account Agreement or Prime Brokerage Agreement with a prime brokerage client will include blanket consent to this practice unless stated otherwise. In general, hedge
funds pay less for the services of the prime broker if their collateral is allowed to be rehypothecated.
There has been very little research in this area. One of the first papers on this topic showed how the collapse in rehypothecation levels was contributing to global deleveraging after Lehman’s demise (Singh and Aitken, 2009a). Adrian and Shin (2009) provide an analytical model where collateral assets can be recycled by pledging and re-pledging; the model shows that during a crisis, the cumulative haircuts (or ‘margin spiral’) on pledged collateral can be sizable. Gorton (2009) shows that during a crisis, haircuts on collateral can result in a run on
the shadow banking system. Singh and Aitken (2009b) show that counterparty risk during and in the aftermath of the recent crisis resulted in a decrease of up to $5 trillion in highgrade collateral due to reduced rehypothecation, decreased securities lending activities and the hoarding of unencumbered collateral.
——–endquote——-
6. April 2012 at 08:39
ssumner:
If a country has a high NGDP growth target, it’s nominal interest rate will never fall to zero, the base will be more than 90% currency, and shadow banking crises will never significantly impact base velocity. I also believe that to be the case when you have 5% NGDP growth rate targeting, as long as it is level targeting, but I can’t be sure.
Well that theory can easily be refuted by simply referring to what happened to the base money to aggregate money stock ratio, or base money to NGDP ratio, from 2008 onwards. Base money skyrocketed, and yet aggregate money M3 and NGDP actually fell. If the Fed were NGDP targeting, then base money would have course been even higher, but not one for one, since even this gigantic increase in base money still didn’t prevent M3 and NGDP from falling.
That shows historical precedent for the possibility that the ratio between base money and the aggregates, like M3 or NGDP, can significantly change.
Why does this happen? One could argue it’s because the Fed is paying interest on reserves. In fact, that is probably what happened in 2008. But, BUT, one could also argue that it’s because the shadow banking system has a far greater role in determining the aggregates like M3 and NGDP, than is typically realized.
I will argue that NGDP targeting carries along with it the potential for the shadow banking system to OVERRULE the Fed’s easy money, if a substantial enough portion of NGDP is composed of shadow banking money. The banks can reduce their lending and hoard all the money the Fed gives (haha) to them, and the Fed cannot get the banks to lend. The Fed could even buy up every last security the banks want to sell, but the Fed cannot force the banks to lend if the banks think investments across the board will lose money.
This is why NGDP targeting implicitly depends on fiscal policy. Without the government bringing about additional inflation financed sales revenues, investors who are given more money from the Fed, cannot earn more profits on what would otherwise be losing investments. It would be like expecting a store that has insufficient sales revenues, to suddenly become profitable if the owners have more cash to increase investment in the company.
Suppose you lived in an economy with only two investment opportunities, one that will lose 10% and the other that will lose 5%. It won’t matter how much money I print off in my basement and give to you. You won’t invest (spend) until the bad projects are liquidated, changed, altered, fixed, all of which takes time. During that time, NGDP will fall and there is nothing I can do to stop it, short of transcending beyond my traditional role of purchasing government debt from my friends in the banks, and buying up the private companies myself and becoming an owner of companies.
6. April 2012 at 08:48
Bill Woolsey:
“Still, I doubt that some rule like the federal funds rate equals .5 times the nominal GDP gap will be robust against changes in shadow banking, and so if someone things that we need such a rule to make the nominal GDP level target credible, shadow banking is still a problem.”
Which is exactly the rule Scott proposes for the NGDP futures payoffs. So the shadow banking is a problem for Scott’s version of NGDP futures targeting.
6. April 2012 at 08:51
“Some economists sound like crybabies””complaining that the zero bound is a big problem and also complaining that we can’t raise the inflation target because the central bank would lose credibility and all sorts of bad things would happen. Fine, but then your two choices are either printing lots of money or suffering a deep recession and debt crisis. Take your pick.”–Scott Sumner.
Excellent, excellent, excellent, excellent. This is how a Market Monetarist should wade into battle.
Yes, you see also that at any time the USA can go into hyperinflation, just like Weimar Republic, with attendant Nazism as the inevitable result. The long, long stretches of moderate inflation and real GDP growth the USA has had—oh, for example from 1982 to 2007 to cite the distant past—were anomalies. It is better we copy Japan. :0
Not to all economists: Read Scott Sumner and then flatter him, and imitation is the best flattery of all.
6. April 2012 at 09:03
Scott,
Currency in circulation has been growing at about a 8-9% y-o-y rate for the past year or so. Significance?
6. April 2012 at 09:35
Rehypothecation is an important dynamic – it impacts the risk associated with asset classes. In a manner, it turns normal assets into a derivative of a base asset (collateral).
The question of whether we SHOULD allow rehypothecation is similar to the question of how much capital should banks be required to hold. Collateral can conceivably support multiple liabilities, as long as the potential losses are lower. Same with bank capital.
So, if a bank or entity pledges collateral 3 times, then so long as the value of the assets does not decline by more than 1/3 (and/or the value of the collateral does not decline), the system is OK. Models of individual risk, however, do NOT do a good job of addressing correlations created by system-wide behavior (like short squeezes on asset classes).
Rehypothecation is central to the understanding of credit money, and it creates what I view as a problem – and what you view as a solution – for the system. If the monetary authority pledges to hit an NGDP target, this will increase rehypothecation as people perceive less price instability in asset classes, and thus seek to obtain cheap/high leverage by posting collateral multiple times. This increases the money supply, which amplifies central bank action. It works like endogenous money.
However, THIS VERY ACTION will create systemic instability by increasing leverage levels (essentially, the same thing as permitting lower capital asset bases).
Central banks seem to LOVE The notion of endogenous money (it creates new money that is NOT their liability!!!). However, private actors not so much… Now, they are forced to create new money and be liable for it.
We can blame the govt. for this – the private sector surely would regulate itself if we had free banking. SURELY. Or we can blame lack of government for this (poor regulation combined with TBTF = moral hazard).
Either endpoint – a high regulation/low leverage system OR a low regulation/high leverage system (with no govt. backstop) might be stable (depending on the empirical evidence), but we’re right in between.
Welcome to no man’s land. Population – us.
6. April 2012 at 09:36
Benjamin Cole:
It is better we copy Japan.
That seems to be more optimal than what the Fed is doing, yes.
Japan has gradually grown in real terms since 1980.
In terms of absolute growth from 1987 to 2007, i.e. the dark blue bars, Japan more or less tied the Euro area, France, Switzerland, and Iceland, and they outperformed Italy, New Zealand, Czech Republic, Portugal, Slovakia, Hungary, Poland, Mexico, and Turkey.
So much for the Japanese lost decades myth.
PS For those who can’t read charts properly, the dark blue bars represent growth in real GDP per capita (adjusted for PPP). The longer the bar, the more growth the country experienced from 1987 to 2007.
PPS For those who want to belittle Japan’s performance because they falsify the myth that real growth requires “loose money”, please understand that I am not saying Japan is a pure and perfect model of monetary policy. So yes, you can point to other countries that surpassed it in terms of performance. My only point is that Japan outperformed many countries that nobody is saying experienced any “lost decades”. Why? Because most economists conflate falling prices with depression, and those other countries had higher nominal statistics over time, unadjusted for PPP or real GDP per capita, so the statisticians believe the rising prices signifies that everything is hunky dory.
Japan outperformed many countries that had “loose money” central banks. That is enough to refute the lost decades myth.
It’s a bitter pill to swallow for those intellectually and politically invested in the “real growth requires inflation, and no inflation makes real growth impossible” ideology. But it’s a pill they must swallow if they are going to consider themselves capable of being convinced through evidence.
6. April 2012 at 09:57
StatsGuy:
Rehypothecation is central to the understanding of credit money, and it creates what I view as a problem – and what you view as a solution – for the system. If the monetary authority pledges to hit an NGDP target, this will increase rehypothecation as people perceive less price instability in asset classes, and thus seek to obtain cheap/high leverage by posting collateral multiple times. This increases the money supply, which amplifies central bank action. It works like endogenous money.
However, THIS VERY ACTION will create systemic instability by increasing leverage levels (essentially, the same thing as permitting lower capital asset bases).
Central banks seem to LOVE The notion of endogenous money (it creates new money that is NOT their liability!!!). However, private actors not so much… Now, they are forced to create new money and be liable for it.
Wow StatsGuy, I am surprised that someone on this board actually understands some of the more important dynamics of inflation.
I think you just touched upon a reason for why Australia has an exponentially increasing aggregate money stock.
One of the most important lessons that economists have to learn is that forcing “stability” in a particular statistic through inflation, be it prices, or interest rates, or NGDP, all of these programs, since that attack and go against the market process of individual private property owners engaging in voluntary exchanges (and lack of exchanges which is just as important), and seeks to impose an artificial magical number that is to be made rigid despite what the collective actions of individuals would result in without the inflation agency, this will invariably lead to a blowing up of other statistics that are not controlled. In the cases of price level and NGDP targeting, the statistic that blows up is cash balances.
The notion that “stable spending” is “optimal”, is a chimera. It has a shelf life. Once the years of inflation distorts the real economy so much that any given quantity of monetary inflation can no longer sustain the distorted economy, then it’s game over for NGDP targeting.
The free market simply cannot be centrally planned to be more optimal than the free market itself. Like white blood cells purging bacteria from the body, the market will eventually purge the economy from the cancerous money counterfeiters. The age old desire of turning stone into bread, of lead into gold, of toilet paper into real wealth, has been around for millennia, and it has never worked, and it will never work. Humans are by nature inimical to being controlled. We are actors. We learn. No amount of constant statistic targeting from central planners can last. They are all doomed to fail, because they are all in violation of what it is that makes us human. Sometimes it takes a few years, sometimes it takes centuries, sometimes it takes thousands of years, depending on how strong the central planning is. But they will fail.
NGDP targeting, if adopted, is going to fail just like price targeting fails. There is no substantive difference between them. They both attack and attempt to overcome the effects of the very important phenomena of voluntary cash holding. They both seek to impose stability in an arbitrary statistic that should only be stable if the collective result of individual action makes it stable. They both seek to overrule the market process. They both seek to control people’s economic lives to some degree.
Market monetarists are not proposing a lasting solution. They are only proposing a new religion to replace an older religion.
6. April 2012 at 10:09
Scott, you wrote:
“I also don’t put much weight on demand deposits. The key is base money and base velocity.”
and
“I’d still say it’s base money that matters, because that’s what the Fed controls.”
I’ve always thought of the banking system as a component of the institutional structure of the monetary system. So these statements are challenging for me.
Is it your view that the government shouldn’t be in the business of chartering banks at all (apart from the central bank)? In other words, the issuance of deposit obligations shouldn’t be a legal privilege requiring a special license?
I’d love to see you write a post with the title “On the irrelevance of depository banks” …
6. April 2012 at 10:28
Separate, but dangerous topic –
http://www.bloomberg.com/news/2012-04-05/end-double-mandate-to-save-fed-s-independence.html
Zingales writes:
“More recently, the Fed’s second round of quantitative easing was a gift to wealthy borrowers with good credit ratings that excluded the deserving needy. To be sure, all these interventions were well intentioned and some were beneficial to the economy. But so are many of the Chinese leaders’ decisions; that doesn’t make them legitimate in a democratic system.”
Suddenly, Zingales’ economic recommendations are based on what creates legitimacy in a democratic system??? Because, that’s what his previous policy recommendations used as a criteria… And, of course, he’s a noted export on democratic theory.
For the record, we (too) should be in favor of ending the dual mandate. We should have an NGDP target only.
6. April 2012 at 10:33
“he’s arguing that near monies matter if you are targeting M2. He’s right’
They also matter it you are attempting to target the price level …
6. April 2012 at 11:23
Scott,
You said;
” I see the hot potato effect as the transmission mechanism that causes spending to rise. But asset prices can certainly affect V.”
I think you are mistaken on this. Generally, if people have more MB than they need, they don’t spend if (I.e. exchange it for real goods and services). They exchange it for other financial assets. What impacts spending is the price (or expected real return, which is just another way to express price) of financial assets generally relative to the price of real goods and services.
What impacts velocity is the spread between the expected real return on money and the expected real return (i.e. price) on other assets.
There is a very strong correlation, which is what creates the confusion on causality.
6. April 2012 at 11:25
Scott,
My comment should read,
What impacts velocity is the spread between the expected real return on money and the expected real return (i.e. price) on other FINANCIAL assets.
6. April 2012 at 11:36
@ Major Freedom
“The free market simply cannot be centrally planned to be more optimal than the free market itself.”
The assumption here is that the current market is a free market, or that free markets can even exist in reality. For a market to exist, a civil/legal/social framework must exist, and this framework is often impossible in a pure free market.
“NGDP targeting, if adopted, is going to fail just like price targeting fails.”
It will only fail if we simultaneously fail to regulate leverage. MTM folks love the fact that markets anticipate CB action, making CB action often unnecessary. Limiting this market response through regulation on leverage would require CB actions of greater magnitude precisely because it would limit financial instability (due to leverage restrictions). However, CBs like the APPEARANCE of market neutrality and therefore of not being political – the illusion of economic technocracy. What happened in 2008 was that everyone saw the emporer had no clothes, that CBs were inherently political institutions with inherently distributional instruments and CBs were forced to take real action (which they delayed, painfully, hoping they could avoid it by bluffing).
NGDP targeting is without question a superior policy. However, NGDP targeting without decent regulation will simply push the economy closer to the redline before the transmission finally breaks for good.
NGDP targeting + quality financial leverage regulation = perpetual AD stability with contained risk
6. April 2012 at 11:42
Stats,
“The first step would be the elimination of the double mandate. Unlike the European Central Bank, which is in charge only of price stability, the Fed has two main legislated goals: promoting full employment and promoting stable prices.
This gives the Fed too much flexibility, pushing it to substitute for the government in designing economic policy. The temptation to act in this way is particularly strong when Congress is divided and paralyzed. It is precisely this substitution that makes the Fed politically vulnerable. The central bank can be independent or activist; it cannot be both. Independent is better.”
6. April 2012 at 12:12
@StatsGuy:
It will only fail if we simultaneously fail to regulate leverage.
I agree 100% we should regulate leverage, however I see regulation and NGDP targeting as somewhat orthogonal (supply vs demand?).
The flip side of debt is that someone bought/invested in something (a house, a factory, etc) expecting to pay it off over N years. Therefore the flip side of debt growth is investment/consumption (ngdp) growth.
regulation impacts the risk-adjusted price of financing (supply side: real return, or potential growth rate) whereas NGDP targeting is more demand side management (in my mind). so crappy regulation means we have a lower ngdp path (its probably a parabola, too little regs and too much regs both imply low potential growth).
NGDP is the best demand-side management tool, but that is orthogonal to what is the best supply-side management tool.
6. April 2012 at 12:42
@Morgan
Yes, I read the argument – but Zingales pulls a switcheroo. He argues we should have only one target to avoid the appearance of politicization, and implies the dual mandate (e.g. full employment) must go.
He gives no consideration to other single targets.
6. April 2012 at 14:37
Patrick, Thanks for that info.
MF, No, base money would have been much lower if we’d been targeting NGDP.
123, That’s completely false. Interest rates play no role in my NGDP rule. I favor targeting NGDP futures.
Thanks Ben.
David Pearson, Exactly what you’d expect with nominal yields near zero. The opportunity cost of holding cash is now very low. As I said in the post, if you do a NGDP target so low that rates fall to zero, you’ll have to print lots of money.
Statsguy, I don’t think there’s much evidence that 5% NGDP growth leads to stable asset prices—look at asset prices during 1982-2007. Admittedly creating a Great Depression will make assets even more unstable, but no one favors that.
MR, You said;
“I’d love to see you write a post with the title “On the irrelevance of depository banks” …”
They are relevant for only two reasons. Government created moral hazard, and lack of NGDP targeting. If we do NGDPLT, and get rid of FDIC and TBTF, they’d I’d go for laissez-faire banking.
Statsguy, In my new post I call for removing the Fed from the process of implementing monetary policy–turn it over to the markets. The Fed should merely set the target (hopefully NGDPLT.)
Greg, I don’t agree on the price level targeting.
dtoh, You said;
“I think you are mistaken on this. Generally, if people have more MB than they need, they don’t spend if (I.e. exchange it for real goods and services). They exchange it for other financial assets.”
You are still missing the basic point. It is IMPOSSIBLE for the public to exchange currency for financial assets (in aggregate.) That’s the insight that all of monetary theory, and damn near all of macroeconomics, is based on. If that’s wrong, all hell breaks loose. Even the MMTers might be right. I’d just quit economics and go to an island somewhere and lie on the beach.
6. April 2012 at 14:56
“That’s completely false. Interest rates play no role in my NGDP rule. I favor targeting NGDP futures.”
Person A holds a bond whose interest is calculated with reference to NGDP gap.
Person B holds the margin deposit at the Fed plus NGDP future with a payoff calculated with reference to NGDP gap.
A and B hold portfolios with identical payoffs. The payoff of NGDP future is economically equivalent to interest.
6. April 2012 at 16:13
123:
“Person A holds a bond whose interest is calculated with reference to NGDP gap.
Person B holds the margin deposit at the Fed plus NGDP future with a payoff calculated with reference to NGDP gap.
A and B hold portfolios with identical payoffs. The payoff of NGDP future is economically equivalent to interest”
You have to explain better.
I don’t think that the overnight interbank loan rate has to equal the expected payoff on the index futures contract.
Since I expect the expected payoff to be zero and the interest rate on overnight loans to be greater than zero, I find it difficult see any obvioius equality.
6. April 2012 at 20:49
Scott wrote:
“PPS. Perhaps before I arrogantly asserted that “rehypothecation” doesn’t matter, I should have looked up the word to find out what it means. But then I’m a hedgehog and Tyler’s a fox.”
I guess according to Archilochus I’m a Fox. But I’ve always been interested in the opinions of Hedgehogs.
7. April 2012 at 00:46
Bill Woolsey,
I was not referring to the interbank loan market. I was referring to how the payouts on central bank liabilities are calculated.
I was comparing two different versions of central bank liabilities.
Version A has IOR set according a mechanical Taylor like rule that uses the current NGDP gap as an input instead of the output gap and inflation.
Version B is Scott’s futures targeting scheme, where the central bank provides margin deposit and NGDP futures combinations to market participants.
Versions A and B have identical payouts, if you set the coefficient in the Taylor like rule so the payout sensitivity to NGDP gaps is the same in both versions.
If you run versions A and B using the same auction process, you have identical macroeconomic results.
This has the real world implications too. If you assume that the ECB is setting the IOR according to the price level gap – quite a reasonable modelling assumption, then the ECB was running the 3 year CPI futures targeting regime during the period of December 2011 – March 2012. Interestingly, the politicians are calling for the reinstatement of the 3 year LTRO auctions.
7. April 2012 at 06:25
Scott,
“Government created moral hazard, and lack of NGDP targeting. If we do NGDPLT, and get rid of FDIC and TBTF, they’d I’d go for laissez-faire banking.”
I think the problem with that is that normal people need a place to put their short term cash for immediate bills where there is no counterparty risk. How many people are sophisticated enough to evaluate the soundness of their bank? I don’t think I am.
We need to get to the place where deposit banking, electronic bill pay, etc., is just the same as paper US currency. Then when people want to invest their money and put it at risk, they buy other assets (stocks, bonds, funds) that are not backed by the government. So keep FDIC for bank deposits, require full reserve of Fed deposits or Treasuries, and then the FDIC would really only insuring depositors against outright fraud.
7. April 2012 at 07:19
123:
I think you are assuming zero interest on 100% margin accounts. Suppose the margin accounts pay competitive interest rates and the percent margin requirement approaches zero?
Also, suppose nominal GDP is expected to be below target? Think about what this implies about the interest return on margin accounts. They are positive and increasing with the size of the deviation.
I think you will find that index futures convertibility is not the same thing as interest rate targeting.
Macroeconomic equivalency? Well, if an interest targeting scheme worked perfectly, so todays interest rate is set by open market operations such that nominal GDP is expected to be on target, or if some kind of index futurers convertility workers perfectly, so that open market operations keep nominal GDP on target, then perhaps it must be the case that the overnight interest rate would be the same.
But this has nothing to do with the pay off on interest futures contract relative to what must be pledged as a margin account.
Further, in an imperfect real world they would not be the same. Index futures targeting would allow for more variation in overnight interest rates than would a system of overnight interest rate targeting.
7. April 2012 at 09:24
Here is a long post on index futures targeting and interest rate targeting.
http://monetaryfreedom-billwoolsey.blogspot.com/2012/04/does-index-futures-targeting-imply.html
7. April 2012 at 10:58
ssumner:
MF, No, base money would have been much lower if we’d been targeting NGDP.
No, that’s false. Base money would have had to have been higher, not lower. NGDP fell post 2008. The only way the Fed could have prevented NGDP from falling, would be if the Fed created even more money than they actually did.
That would have made base money rise even more than it did.
7. April 2012 at 11:07
StatsGuy:
“The free market simply cannot be centrally planned to be more optimal than the free market itself.”
The assumption here is that the current market is a free market, or that free markets can even exist in reality.
Not the former, but definitely the latter. Of course free markets can exist in reality. All it takes is people choosing it, the same way totalitarianism can exist in reality, by people choosing it.
For a market to exist, a civil/legal/social framework must exist, and this framework is often impossible in a pure free market.
I like your “often”. It means you accept that a civil/legal/social framework is indeed possible.
Saying “often impossible” is an oxymoron. Impossible means it can never occur. Often impossible is like saying sometimes never.
“NGDP targeting, if adopted, is going to fail just like price targeting fails.”
It will only fail if we simultaneously fail to regulate leverage.
No, it will fail even if you find omniscient angels who will “regulate” leverage. Regulate it how? By how much? Where? What rules? It’s all garbage.
MTM folks love the fact that markets anticipate CB action, making CB action often unnecessary. Limiting this market response through regulation on leverage would require CB actions of greater magnitude precisely because it would limit financial instability (due to leverage restrictions).
If market correctly anticipate CB actions, then the CB MUST act if the market is going to be considered correct in their expectations.
It’s a contradiction to say on the one hand that the market correctly anticipates CB action, and then on the other hand you say the CB doesn’t act at all.
However, CBs like the APPEARANCE of market neutrality and therefore of not being political – the illusion of economic technocracy. What happened in 2008 was that everyone saw the emporer had no clothes, that CBs were inherently political institutions with inherently distributional instruments and CBs were forced to take real action (which they delayed, painfully, hoping they could avoid it by bluffing).
Thank the relatively free internet.
NGDP targeting is without question a superior policy.
It will fail.
However, NGDP targeting without decent regulation will simply push the economy closer to the redline before the transmission finally breaks for good.
“Decent regulation” means “I don’t know how it will work, but men with guns can save us all.”
NGDP targeting + quality financial leverage regulation = perpetual AD stability with contained risk
Two eyes of newt, 20 ounces of dragon blood, 4 bat wings, and 3 fairy tears. Bring to a boil, then chant “abracadabra.”
7. April 2012 at 11:26
123, That’s completely wrong. My proposal doesn’t even require the existence of interest rates, it would work fine in an economy with no credit at all, so it’s obviously not dependent on interest rates. I have not proposed the creation of bonds indexed to NGDP.
Thanks Mark,
Negation. I envision a system where any FDIC deposits need to be backed 100% with government bonds. That would completely remove moral hazard from the system. Those who wanted to shoot for higher rates of return on their investments, could invest in bank deposit not backed by FDIC, or equities.
That’s what I meant, I don’t have time to spell out the entire proposal each time, and sometimes just say end FDIC to save time.
MF, I must have explained that 100 times. I’m done.
7. April 2012 at 11:59
ssumner:
MF, I must have explained that 100 times. I’m done.
You were wrong 100 times. No individual investor or seller cares a lick about NGDP. They only care about their own sales and their own incomes.
Individual market actors are not even able to plan around NGDP changes.
If the Fed commits to maintaining NGDP, then it’s not the case that individual actors will require fewer dollars before they decide to spend, such that base money will be smaller.
The charts you linked to that compared country NGDP and base money, were fallacious comparisons, because the correct comparison is the counter-factuals in each country that you cannot observe. In each example, a higher NGDP would have required higher inflation from the central bank, i.e. more base money.
You kept comparing base to GDP, and say that NGDP targeting would have a lower base to GDP ratio, when I was referring to only base.
8. April 2012 at 00:32
Scott,
your proposal requires margin accounts or other means of establishing creditworthiness.
Your “economy without credit” is just an economy where the central bank is the sole provider of credit.
“I have not proposed the creation of bonds indexed to NGDP.”
You have proposed the equivalent. NGDP future plus a margin account is a NGDP indexed bond. By definition.
It is just that for convenience NGDP indexed bonds with high sensitivity to changes to NGDP are called NGDP futures, and NGDP futures with high margin requirements are called NGDP linked bonds. And it is trivially easy for the users of such instruments to transform one into other by adding or subtracting the regular bonds to their portfolios.
8. April 2012 at 04:54
Bill Woolsey:
“I think you are assuming zero interest on 100% margin accounts.”
Well, I wrote about the more general case, with a fixed interest rate on a fixed percentage margin account. For example, 3% interest rate and 10% margin account.
“Suppose the margin accounts pay competitive interest rates and the percent margin requirement approaches zero?”
When the margin percent requirement approaches zero, the probability of central bank default approaches 100%.
How do you know what is the competitive interest rate? Any deviation from the optimal interest rate on margin accounts will distort the NGDP futures peg.
More comments will follow during the next couple of days. This all is a very interesting intellectual challenge.
8. April 2012 at 14:44
MF, If everyone individually cares about their own sales, then collectively the market cares about NGDP.
123, If you want to call a futures contract a “bond” that’s fine, but I don’t see any substantive implications. The payoff isn’t even positively related to NGDP (if you go short.)
8. April 2012 at 15:32
MF, If everyone individually cares about their own sales, then collectively the market cares about NGDP.
That’s the fallacy of composition.
Each individual caring about their own individual sales is the only “caring” going on. “The market” is not doing any caring, since “the market” is not an individual.
NGDP can go up or down, or stay the same, and to the individual, it is irrelevant. Individual sales can go up or down, or stay the same, and to the individual, it is relevant.
There is no such thing as “collectively” caring. There is only individual caring.
Since the individual does not care about NGDP, targeting NGDP is not only a vain quest, but as mentioned many times before, it is positively destructive since it distorts economic calculation due to the fact that targeting NGDP requires inflation and messes up interest rates.
9. April 2012 at 01:44
Scott,
“If you want to call a futures contract a “bond” that’s fine, but I don’t see any substantive implications.”
Well, Bill Woolsey saw some substantive implications. See his resent posts “Does index futures targeting imply a target for the interest rate?” and “Convertibility with nominal GDP Indexed Bonds”.
See my reply to Bill here:
http://monetaryfreedom-billwoolsey.blogspot.com/2012/04/does-index-futures-targeting-imply.html?showComment=1333915826679#c5596698740660190002
Here is Bill:
“Interestingly, the interest on margin accounts does seem to provide an equilibrium condition if nominal GDP is expected to be “too high.””
Bill goes on to argue that since the interest rate on margin is set with the reference to one year T-bill rate, the Fed is able to avoid pegging the interest rate.
Let’s examine the Mankiw rule:
Federal funds rate = 8.5 + 1.4 (Core inflation – Unemployment)
So in effect the you and Bill are making two criticisms of the Mankiw rule:
The first is that the Mankiw rule should be changed to:
Federal funds rate = 1 year T-bill rate + 3.5 + 1.4 (Core inflation – Unemployment).
The second criticism is that the 3.5 + 1.4 (Core inflation – Unemployment) term should be replaced with the NGDP gap.
I agree with the second criticism. I am not so sure about the first.
What if the 1 year T-bill rate used as a reference for margin account interest is not available (in the Eurozone there are no Eurozone T-bills)?
Another problem. What if the 1 year T-bill rate used as a reference for margin account interest is too high or too low? For example, after an overnight shock in Asia, markets start expecting that NGDP will deviate from target. As a result, 1 year T-bill rate is 0.5% lower than it would be if NGDP expectations were on target. Then the parameters of the 12:00 NGDP futures auction are wrong. The auction will result in a wrong quantity of base money being created or destroyed.
9. April 2012 at 01:49
Scott: “The payoff isn’t even positively related to NGDP (if you go short.)”
For every short there is a long. So for every future, there is a party to the contract that receives the payoff positively related to NGDP.
9. April 2012 at 03:30
Hi Scott,
since this point is where I differ from your worldview, could you please adress the mechanism in following sentence:
“If a country has a high NGDP growth target, it’s nominal interest rate will never fall to zero, the base will be more than 90% currency, and shadow banking crises will never significantly impact base velocity.”
I understand the first part, but I miss the move from 90% currency to insignificant shadow banking. I suspect you have in mind that since most of base money is currency, shadow banking affecting the rest 10% is irrelevant. But what is your transmission mechanism – is it just hot potato? and if yes, how do you expect the money to come to crculation if banks will not make new loans/deposits?
Thans,
CS
10. April 2012 at 11:50
123, I don’t see any relationship to the Mankiw rule, which is backward looking, isn’t it? My proposal is forward looking.
I don’t think T-bill yields need be used at all as a reference interest rate. Just use my subsidy proposal to encourage trading. The subsidy per trade is the total subsidy divided by the number of trades. Then the market determines the interest rate on margin accounts.
I agree that there are roughly equal long and short positions, but don’t see why that matters. In aggregate, the public is indifferent.
CS, If interest rates are positive then banks don’t want to hold excess reserves. If there are no good loan opportunities, then they buy safe government securities, and that puts the reserves into circulation.
10. April 2012 at 16:05
Scott:
Exactly how do margin accounts work in your system?
123:
If the central bank buys futures contracts, what exactly do you see happening?
What exactly happens with margin accounts?
10. April 2012 at 16:05
Scott:
Exactly how do margin accounts work in your system?
123:
If the central bank buys futures contracts, what exactly do you see happening?
What exactly happens with margin accounts?
10. April 2012 at 17:44
Is there any reason not to run it first like a Vegas Sports Book and then afterward like a Central Bank?
1. Anyone funds acct. and bets.
2. The betting predicts NGDP up until the monthly (soon after weekly?) NGDP number is published, until that time the odds changes with bets.
3. This money is taken off the table first and new printed money gets paid out here first.
4. Anti-house caveat: instead of Forex model, where you are paying vig, instead there is no spread. In fact, betters lose a little teeny bit less than they should, and win a little more than they should.
5. Then based on outcome Fed does other stuff they normally do helping GS get rich.
Now the first node where money enters or leaves system is with the betting public.
The Fed has a nice prediction market running on as many months years out as people want.
AND since there is a very slight finger on scale, Fed is clearing favoring this Vegas / Forex system, and removing any suspicion that the Fed props up US debt.
11. April 2012 at 08:42
“I don’t see any relationship to the Mankiw rule, which is backward looking, isn’t it? My proposal is forward looking”
Futures are backward looking when they expire. So it is possible to compare futures vs. Taylor/Mankiw rules. And Taylor is as angry about Bernanke changing his coefficients as you would be about the law that changes the payoffs of the existing NGDP futures.
“I agree that there are roughly equal long and short positions, but don’t see why that matters.”
I meant that for each single contract, long position of one of the parties exactly matches the position of the short party. By definition. So you can always compare the payoffs of the long-NGDP party to the Taylor rule.
11. April 2012 at 09:08
Bill,
In my system the long and short positions of the central bank are exactly matched.
The Fed is borrowing when it sells a nominal GDP index bond to a bank or firm. The nominal GDP indexed bonds that the Fed sells are its own monetary liabilities. The interest is calculated as 4.5% plus X times the current NGDP gap. There are no other liabilities. There are no margin accounts for monetary liabilities.
The Fed is lending when it buys a nominal GDP index bond from a bank or firm. The nominal GDP indexed bonds that the Fed buys are assets to the Fed. When the Fed buys the bonds, it pays for them by selling monetary liabilities (4,5% plus bonds). The Fed receives 5,5% plus X times the current NGDP gap. There are no margins, however a bank or firm has to post collateral (for example corporate investment grade bonds).
On a daily basis, the collateral of nominal GDP indexed bonds that the Fed holds is marked to market with a haircut. All the returns from the collateral accrue to a bank or a firm. If the market value of the collateral has increased, some part of the collateral is released. If the market value of the collateral has decreased, bank or firm has to post additional collateral, or else the Fed repossesses and sells the collateral.
The Fed is passive and expands or contracts its balance sheet according to the fluctuating demand for money that pays 4,5% plus X times NGDP gap and fluctuating demand for safe overcollateralized loans that yield 5,5% plus X times NGDP gap. All the interest is paid daily.
12. April 2012 at 11:52
Bill, They are strictly to prevent losses to the Fed if bettors default. People put down a modest percentage in a margin account, and enough interest is paid on the margin account to assure a highly liquid NGDP market.
Morgan, You could try it out first as a pure gambling market, then gradually link it to Fed policy.
123, You said;
“Futures are backward looking when they expire.”
Sorry, but I don’t even know what that means. Expiration has no impact on monetary policy.
12. April 2012 at 12:38
Scott, suppose Taylor Rule with NGDP gaps is enshrined in law. This means that Fed Funds Rate future is a NGDP future (do the math). Note that FFR futures do exist and are actively traded in Chicago.
So if the Taylor rule is credible (for example if 3 year FFR contracts trade on NGDP target), deviating from it makes no sense, as it is in effect equivalent to leaving NGDP convertibility regime. That’s why Taylor is so angry about slight deviations from his rule during the housing boom period.
“People put down a modest percentage in a margin account, and enough interest is paid on the margin account to assure a highly liquid NGDP market.”
This is too circular. Suppose the liquidity in NGDP market has doubled. Do you reduce the margin interest rate? Any decision will have a big impact on the quantity of base money.