Belated write-up of AEA/AFA meeting, Part I
Update: I accidentally posted an earlier draft of this post rather than the intended final draft. I’ve restored the intended final draft. Changes are listed at the end of the post.
Two weeks ago I had great fun in Denver spending three-days attending seminars and lectures given by the great and good of academic economics. Since then, I’ve been holding it all in my head precariously, until I write some things up. (Writing is the process whereby I permit myself the pleasure of forgetting.) I expect that what follows these words will be long, sprawling, and disorganized. But for what it’s worth.
A note — The AFA gave me a $1500 grant to attend. That was kind of them. Thanks.
General Impressions
This was the first time I’d attended an economics conference. But in my early adolescence, I did frequent another sort of hotel-bound gathering, and the resemblance was uncanny. The AEA is basically a Star Trek convention in suits. It’s a gathering of the same sort of geeks. The same combination of earnestness and awkwardness marks off and distinguishes the attendees from normal business travelers. Star Trek conventions have their celebrities, here’s George Takei, there’s Nichelle Nichols. The AEA has its celebrities as well: the Nobelists, the famous economists from Chicago, Harvard, and MIT whose papers you have read (or you pretend to have read). Like a kid at a Star Trek convention, I had great fun at the meetings. Still, there were undercurrents that made this affair feel less innocent — so many PhD students nervously interviewing for jobs; the networking and earnest introductions; the faint, polite stench of status competition. At a Star Trek convention, everyone wants to meet George Takei. No one is trying to become him.
The Highlight
For me, the highlight of the meeting by far was lunch with Scott Sumner and Scott Wentland. We had a grand conversation. Readers of both blogs might imagine the authors of The Money Illusion and interfluidity to be on opposite sides of a great divide, but it didn’t feel like that at all. The quality of mind I value in other people and strive for in myself is a kind of nimbleness, a fluidity of mind. The world is too complex for any particular narrative to be perfect. Good judgment, I think, comes from the ability to slip between and among stories, to understand the ways different accounts might be true, to marshall evidence and reasoning on both sides and then assign weights to a superposition of competing, sometimes contradictory ideas, all of which play a role in ones choices. Sumner and I understood our different perspectives very quickly, and took one another seriously, though we’d probably weight accounts very differently. Further, though I suspect he will bristle a bit at the characterization, within the economics profession I view Sumner as an ideologue in the very best sense. There’s both a moral and a methodological component to that. Sumner is driven, scandalized even, by what he sees as a profound and preventable failure of monetary policy. He’s shocked that the rest of his profession (which he’d previously considered himself to be in the middle of) has shrugged this off, that economists don’t get in their guts how awful an abdication of policy has occurred. So Sumner has made it his full-time preoccupation for two years to communicate and persuade, working to change his colleagues’ intuitions about what is acceptable and what is not. He has a reasonable (though not unassailable) model of how the economy works, and a coherent vision of a policy regime that would be wise under that model. Recent experience suggests that implementing Sumner’s policy regime, under which the monetary authority both commits to and is able to target NGDP, would be eased by tools that are institutionally or politically unavailable under current arrangements (e.g. NGDP futures markets, negative interest on reserves, perhaps more flexibility with respect to asset purchases). Rather than working within existing constraints, he has made lobbying to alter them part and parcel of his campaign to shift the intuitions of his colleagues with respect to the conduct and duties of monetary policy.
I’m not entirely on board with Sumner’s project. I have longstanding concerns about status quo monetary policy. I’m not sure NGDP is a sufficient statistic for a decent economy. I share some of Arnold Kling’s concerns that monetary policy may be unable to solve information problems with respect to patterns of production, consumption, and income, along with old Austrian-ish concerns that monetary expansion can lead to counterproductive distortions towards “dumb” interest-rate sensitive investment. I’m not sure that the Fed credibly could target NGDP, even with the expanded toolkit Sumner proposes, and I worry about the fiscal costs and economic consequences if markets test and manage to break a drifting NGDP peg. Sumner offered some interesting rejoinders. He pointed out that the worst distributional effects of crisis policy — the various bailouts and subsidies intended to put a floor under outcomes for “systematically important” financial institutions, the panicked money-flows post-Lehman — might have been avoided if NGDP-targeting monetary policy were sufficiently credible. If the path of NGDP is certain, it is possible that no institution would be too big to fail. The idea is that, whatever micro-level complications and litigations and reorganizations the failure of a major bank might provoke, if at a macro-level real GDP and employment remain sufficiently OK, nonintervention would become politically and morally thinkable. (Of course, you can argue this is wrong, that big bank failures cascade so disruptively that pegging NGDP would be insufficient to prevent a collapse of real production, so policymakers would continue to intervene. But note the congruence of Sumner’s view and Rajiv Sethi’s.) Sumner dislikes and generally opposed bank rescues, but he pointed out that one way to look at the subsidies to banks is government undoing costs inflicted by bad policy. Nominal debt is contracted around expectations about nominal growth, and by failing in its duty to ratify those expectations, monetary policy failure was responsible for the increased debt burdens and reduced asset values that harmed banks. Therefore, some compensation might be justified. That’s an interesting argument, but it turns on what expectations we deem reasonable ex ante. The Fed has never committed to NGDP level-targeting, so perhaps banks ought to have been expected to manage leverage cautiously and to be tolerant of fluctuations. Moreover, the argument can’t explain or justify the distribution of intervention during the crisis. If government is responsible for changes in the real debt burden associated with failure to stabilize NGDP, then there ought to have been compensation for indebted households and nonfinancial firms. But subsidies and interventions went disproportionately to banks, and disproportionately to just a few banks.
Despite some misgivings, I think Sumner’s project is serious and interesting, and we could do a lot worse. It’s not exactly what I would push, but there’s plenty of overlap and I wish him well. At a high level of abstraction, I find Sumner’s “center right” views to resemble the “far left” post-Keynesian Chartalists, or “MMT-ers”. Both Sumner and the MMT-ers choose a macro target and a policy instrument, and suggest that micro problems will work themselves out if the consolidated government/central-bank is vigilant about supporting the target. MMT-ers choose (net) fiscal spending as their instrument, while Sumner chooses monetary policy under an unconventionally expansive definition. Some MMT-ers would target unemployment (often at zero, via a direct government jobs guarantee). But others argue that the government should deficit-spend at the level that supports GDP without provoking inflation, which is not too different from Sumner’s NGDP target. (Sumner argues that, at reasonable growth rates, NGDP targets are likely to be met by sustaining real GDP rather than by inflation.) Am I alone in seeing the similarities?
Like Andy Harless (but see Sumner’s rejoinder), I think the distinction between fiscal and monetary policy has grown very blurry. Monetary reserves are now interest-bearing obligations, ultimately paid for by the state. Some Fed “liquidity facilities” involved issuing interest-bearing obligations to buy up private sector assets (at prices above those offered in private markets). That sounds like fiscal policy to me. While it can be argued that conventional open-market operations only transform the maturity of government obligations, by anchoring the yield curve and increasing the fraction of debt that can be used directly as a medium of exchange, conventional monetary policy may increase the willingness of private agents to hold US debt, reducing constraints on spending and enabling expansionary fiscal policy. Fiscal policy and monetary policy are intertwined, and it’s not clear to me that either dominates the other. (There’s an aphorism, I think Tyler Cowen’s originally, that “the monetary authority moves last”. That doesn’t persuade me. Timing of endogenous phenomena tells one very little about causality. Timing of moves in a game tells us very little about which player has the advantage.) Ultimately, I’ve come to think that the main differences between fiscal and monetary policy are institutional. Decisions about what we call “fiscal” and “monetary” policy decisions are made in different ways by dissimilar entities. Those decisions can reinforce one another, or they can offset and check one another. Some people prefer to emphasize the role of fiscal authorities for “democratic legitimacy”, while others champion action by an “independent central bank”, on the theory that isolation from overt politics will yield technocratically superior choices. You can accept these preferences on face, or more cynically argue that some groups expect one or the other decisionmaking body to execute policy in ways that that favor preferred interests. Regardless, at a macro level, Sumner’s NGDP targeting monetary policy and MMT-ers’ GDP-supporting fiscal policy look similar to me. Both perspectives arouse my sympathies but provoke misgivings. First, I’m not sure either instrument is up to the task of stabilizing the target over a long horizon, and worry that attempting but failing to stabilize may prove riskier than conventional muddling through. Second, I think the micro-level stuff really does matter. In order to ensure both high quality resource allocation and distributional legitimacy, I think it matters very much what is paid for with fiscal expansion, and precisely how monetary policy is to be conducted. (I offered a proposal a while back that now looks like a bizarre hybrid of Sumnerism and Chartalism, which tries to address micro-level concerns.)
I’ve been remiss in not saying much about Scott Wentland, who was actively engaged in our conversation but is less clearly identifiable with a position. Wentland describes himself as a devil’s advocate, but I’d characterize him more as a satanic Socrates — he’d listen, carefully reinterpret a comment, then politely punctuate his review with a challenging question. Still, for all the finance and economics I encountered at the conference, Wentland is the only person whose work suggested a way to actually turn a profit. Wentland presented a paper at the conference. I missed the presentation, but read the paper after the fact. It is empirical work very nicely done, and it tweaked the antennae of my inner, amoral arbitrageur. I now think of registered sex offenders as roving Groupons for home flippers. Wentland and his coauthors provide strong evidence that you could make a lot of money persuading an ex-cellmate to move near a nice, four bedroom home in rural Virginia, and then to move away after you’ve bought the home.
[Update: Wentland et al simply documented the effect on home prices and liquidity of nearby registered sex offenders, in a careful and empirically sophisticated way. The “arbitrage” is my poor attempt to say something clever about it. However, commenters inform me that the scheme I’m implying mirrors an old and well-known strategy with racial overtones, a parallel which I did not intend to draw. Thanks to commenters Kindred Winecoff and TGGP for pointing this out.]
Anyway, those are my musings on lunch. I’ll take a breather and leave my comments on the lectures and seminars I attended to future posts.
- 23-January-2011, 6:50 p.m. EST: Restored intended final draft — somehow I mistakenly posted an earlier draft! This draft differs from the previously posted draft:
- The current draft makes clear that Scott Sumner generally opposes bank resucues, despite the argument that some bank subsidies can be viewed as compensation for bad policy
- The current draft adds a line re Scott W’s “devil’s advocacy”
- The current draft omits a gratuitous joke re fraud at banks;
- The current draft adds a comparison between Sumner’s views and those of Rajiv Sethi
- The current draft includes more links in general
- Probably some other minor differences
- 23-January-2011, 7:20 p.m. EST: Added update re “blockbusting”, Thanks to commenters Kindred Winecoff and TGGP.
- 25-January-2011, 7:10 p.m. EST: Corrected mssplling of “Chartalists”, many thanks to supercommenter JKH for pointing out the error!
[…] This post was mentioned on Twitter by Stuart Horrex, Conduit Journal. Conduit Journal said: Belated write-up of AEA/AFA meeting, Part I http://bit.ly/i9Vy4c […]
January 23rd, 2011 at 2:41 am PST
link
Steve,
A few comments on MMT as it relates to your post–
1. We prefer targeting total employment (with a labor buffer stock that rises/falls counter to the private sector), not nominal GDP. See http://www.levyinstitute.org/publications/?docid=1349
2. To suggest that micro problems will work themselves does not characterize us at all. True, we think macro policy is the main starting point, but it must be administered appropriately at a micro level (and even our preferred policy of targeting employment doesn’t do this all by itself, we are aware). See here http://e1.newcastle.edu.au/coffee/pubs/wp/2005/05-09.pdf where the authors argue that “the blunt instrument of orthodox Keynesianism . . . fails to take account of the spatial distribution of social disadvantage.” For a bit more empirical support for the argument, see this– http://e1.newcastle.edu.au/coffee/pubs/wp/2005/05-05.pdf
Best,
Scott
January 23rd, 2011 at 9:50 am PST
link
Re: Wetland’s paper: That is an old game, although it used to involve moving in a minority family. The recent film “An Education” shows an instance of this basically as an aside.
January 23rd, 2011 at 11:32 am PST
link
No one should ever listen to someone that can’t see the forest thru the trees. The money supply can never be managed by any attempt to control the cost of credit — so if for no other reason, that competely excludes MMT. Sumner is indeed correct, in that nominal gDp rose at increasing rates-of-change, until its crescendo in the 1st qtr of 2006. (very clearly seen using a 3 qtr rate-of-change). I.e., Greenspan never “tightened” monetary policy – despite raising the FFR on 17 separate occasions – over 41 consecutive months.
But how do you target nominal gDp? The answer is that you use monetary flows, or aggregate monetary purchasing power (our means-of-payment money X’s its transactions rate-of-turnover). I.e., MVt is equal to nominal gDp.
MVt bottomed in Oct 2002. It then rose until Jan 2006. Bernanke then initiated a “tight” monetary policy which he continued for 29 consecutive months, not stopping to “ease” when Bear Sterns 2 hedge funds collapsed, but waiting until Lehman Brothers declared bankruptcy, thereby driving the economy into our depression. But that is money flows proxy – its mirror. Bank debits is the ultimate monetary metric.
January 23rd, 2011 at 12:20 pm PST
link
Sumner’s use of “level targeting the forecast” should me mentioned – which allows the Fed to guarantee that they will “make-up” as quickly as possible for any deviations (or market-tests) and restore the NGDP path to its desired trend-line. I agree his plan is far from unassailable (I think he’s far too dismissive and insouciant about the existence of bubbles fueled by debt-financed speculation and irrational exuberance), but is probably superior to anything we are hearing about now, and avoid the corruption inherent in crisis-based ad-hoc and highly politicized improvisations among friendly insiders conducted under the guide of emergency fiscal stimulus.
You’ll never believe much in the ability of our government to conduct intelligent and honest fiscal stimulus when you see the locally-obtained grant go to the mayor’s brother’s company to rip up perfectly good sidewalks and rebuild them while the collateral road remains pocked with pot holes. Macro-tactics will always tend to be seen as having more integrity, justification, and inherently more technocratic.
The effect of his plan on nominal-debts is particularly intriguing. In a boom with lots of real growth expected inflation falls which discourages more credit accumulation, and in a bust with zero or even negative real growth – inflation rises to rapidly wipe out the real burden of existing debts – relief which continues and compounds until real growth is restored. Taking a look at our old friend Fred and putting it next to even a 4.5% NGDP trend line (Sumner prefers 5%) shows that since the recession we’re still 5.5% cumulative inflation behind the decadal trend. At 5% it would be 10.7%. That would have been a lot of debt relief at just the right time – not to mention the employment effect in a sticky-wages environment.
The downside of low expected inflation in a boom is the while debtors may be more reluctant to take on more debt, creditors may be more willing to extend it. The sides are simply not symmetric in discipline, self-control, sophistication, rationality, and bargaining position – and my major concern is that this imbalance can undue some of the benefit of the proposal. However, it’s nothing a little enforcement of prudent underwriting standards (proof of income, 20% down, fixed-rate mortgages, that kind of thing…) and the removal of unjustifiable explicit or implicit government guarantees and subsidies (Fannie Mae, Freddie Mac, that kind of thing…) couldn’t fix.
January 23rd, 2011 at 3:26 pm PST
link
Don’t tell me Wentland has re-invented block-busting.
January 23rd, 2011 at 6:51 pm PST
link
“(Of course, you can argue this is wrong, that a big bank failure would cascade so disruptively to households and firms that pegging NGDP would be insufficient to prevent a collapse of real production, so policymakers would continue to intervene.)” I’d like to see the argument; surely it would be quite implausible. If real production collapsed though NGDP expectations continued to grow (mildly, at their wonted steady pace), inflation would have to go through the roof. When has such a combination ever taken place?
The main reason a big bank’s failure would make us worry about the whole financial system is our fear that it would occasion a precipitous drop in NGDP, leading (for example) to other bank failures. Sumner’s plan would eliminate that fear.
January 23rd, 2011 at 9:06 pm PST
link
I offered a proposal a while back that now looks like a bizarre hybrid of Sumnerism and Chartalism, which tries to address micro-level concerns.
And I thought it was your finest hour! :o) (“During depressions and disinflations, the Fed should be depositing funds directly in bank accounts at a fast clip. During booms, the rate of transfers should slow to a trickle.”)
A good idea certainty, but fiscal policy can only be set by Congress and instead of a universal cash transfer program, it’d be administratively (and politically) easier to simply tax everyone less. A few years ago, Wynne Godley made a strong case that fiscal policy is the best way to move forward, and considering we’ve been at zero bound for two years, perhaps the only way.
We also show that fiscal policy on its own could achieve both full employment and a target rate of inflation.
http://www.levyinstitute.org/publications/?docid=911
The catch is that the Fed’s discretion to set monetary policy can adjust faster than Congress’s discretion to set fiscal policy, with the limited exception of automatic stabilizer programs like UI and food stamps that pre-authorize Tsy to spend by formula. The way to speed up the decision cycle on fiscal policy is to pre-authorize Tsy to tax by formula.
For example, by a statute that adjusts FICA rate based on DOL’s monthly U3 unemployment report. With 9.4% unemployment, FICA rates reduced by 94% (a 7.0% U3 rate, FICA reduced by 70%, etc.) would be the right order of magnitude. Month by month, FICA rate would automatically adjust (hopefully) upward as economy approaches full employment (hmmm, wasn’t Weintraub’s tax-based incomes policy a wage tax?).
January 24th, 2011 at 9:50 am PST
link
Steve, I think that Tyler’s ‘monetary authority moves last’ refers to the fact that central banks can change interest rates quickly and decisively, while fiscal policy is much more inflexible – so monetary policy ends up determining the overall macro-policy stance.
In other words, a central bank can change monetary policy quickly to ‘cancel out’ any change in fiscal policy that the monetary authorities deem as throwing the system out of the equilibrium they want. The government cannot change fiscal policy as quickly, hence for macro purposes it is as if the monetary authorities always ‘have the last word’ on what the overall policy stance is. Equivalently, ‘the monetary authority moves last’.
January 24th, 2011 at 12:58 pm PST
link
Monetary policy always intersects with fiscal policy. The standard central bank balance sheet is a source of cheap funding (currency) for the treasury (consolidated), which is a fiscal effect. Non-conventional policy expands this to asset-liability credit trades, supported by marginally expensive, interest paying reserve funding (compared to currency).
Sumner’s NGDP proposal relies heavily on monetary policy, obviously. Some people get confused by overlooking the fact that NGDP futures are only one part of the full mechanism – the more critical part in fact is the assumption that quantitative monetary policy will have the intended effect on NGDP that NGDP futures traders are betting it will have – including the ones who are most right. This amounts to the assumption that the effect of quantitative monetary policy on NGDP is a continuous function. It is a big assumption in a world where conventional monetary policy mostly targets interest rate levels before reserve quantities. And of course it is critical to the notion that NGDP futures targeting would have prevented the financial crisis and/or the great recession.
I think there is a similarity with MMT in quite a different way. MMT in effect relies very heavily on monetary policy, although it positions itself nominally the other way around. A central tenet of MMT is that a fiat currency issuer can always afford to run deficits. What is sometimes unsaid however is that this claim relies on the assumption that the currency issuer always has the option of abandoning bond financing, using the mechanism of bank reserve creation to close the loop between crediting bank accounts through spending and issuing state liabilities as a result. That is how the state frees itself from the potential hostage taking threat of the bond market, which otherwise might force its hand. Without that option, the funding capability of the US government for example at a purely technical level would face the same risk TYPE as Greece, to whatever degree, because neither would have full sovereign control of a backup monetary mechanism for fiscal funding. Of course, that option is implicitly built in to a sovereign currency issuer such as the US, but the point is that the option must be exercisable in order for its implications to have any real bite. So the option whereby fiscal policy fully usurps the function of monetary policy is essential to the MMT interpretation of the real world potential for a fiat currency issuer. That in its own way is a fairly heavy reliance on monetary policy.
P.S. – typo re “Post-Keynesian Chartalists”
January 24th, 2011 at 7:25 pm PST
link
This isn’t theory, it is evidence based. “True-ups” are never radical, they are subtle corrections. The rate-of-change in money flows is always measured with the same length of time as the specific economic lag (as its influence approaches its maximum impact (not date range); as demonstrated by the clustering on a scatter plot diagram). These lags were set at closed intervals for the last 97 years. Lags take a long time to complete, but thier path, not long to correct.
A price level target is prefereable to targeting nominal gDp, because limiting (n)gDp would also cap real-output. I’m sure that by using (n)gDp, we could avoid an overheated economy, but that would be overkill.
There is no true-up with MMT, inflation accrues overnight.
January 25th, 2011 at 11:40 am PST
link
JKH,
Completely agree on both counts. NGDP “targets” assumes the New Consensus-like “expectations” are all important. Otherwise, there’s absolutely no transmission mechanism. MMT proposes a central bank that creates (or at least can create) the “space” for fiscal sustainability.
January 25th, 2011 at 12:41 pm PST
link
Scott,
I’m glad that made some sense to you.
SRW,
Another lunch worth having:
TORONTO — Canada’s top financial regulator has made it a “huge priority” to compel the country’s banks to draw up so-called living wills explaining how their operations could be dismantled in the event of failure.
In an interview with the Financial Post, Julie Dickson, the Superintendent of Financial Institutions, said some of the biggest banks have completed the task, part of a broader initiative aimed at demonstrating how the financial system could continue to function without government bailouts in the event of a meltdown.
http://www.financialpost.com/news/financials/Banks+road+OSFI+head/4159405/story.html
January 25th, 2011 at 3:16 pm PST
link
I eagerly await the 2% + inflation premium to be injected into long-term interest rates with the adoption of MMT.
January 26th, 2011 at 4:47 pm PST
link
Based upon monetary flows HOUSING BOTTOMs in JAN. Of course the Case-Shiller index is a 3 month moving average – so its bottom will come a month or two later.
“This index family includes 20 metropolitan area indices and two composite indices as aggregates of the metropolitan areas. The composite and city indices are normalized to have a value of 100 in January 2000”
MVt peaked in Feb 2006. Case-Shiller then peaked in the 2nd qtr 2006 @ 189.93 (because of the 3 month average).
Options and futures based on Case–Shiller index are traded on the Chicago Mercantile Exchange.
January 27th, 2011 at 6:44 pm PST
link
Steve,
I’ve been thinking similar things re Chartalists and Monetarists myself recently: they’re twins, with the Chartalists targeting slightly different but equivalent aggregates (“NFA”). Both schools fall victim to Goodhart’s Law, aka the Lucas Critique.
This also neatly explains the rather weird and aggressive obsession some Chartalists have with Monetarism. They’re brothers, of course they hate each other!
January 31st, 2011 at 8:32 am PST
link
Goodhart’s Law and Lucas Critique are quite different. And Goodhart is a chartalist, or at least a quasi-chartalist.
February 1st, 2011 at 1:27 pm PST
link
And, as I noted in comment 2 above, we don’t target NFA. NFA is the appropriate “monetary aggregate,” in our view, but it’s not something you necessarily target.
February 1st, 2011 at 1:31 pm PST
link
Scott,
Hahaha! Just trolling you mate. That said, fair enough on your second comment. I could have been more careful in what I wrote—I had in mind a target govt deficit defined not in its own terms but in terms of the full-employment level of output or some other long-run equilibrium at which realised net saving was stable (per Solow, quoted in Godley and Lavoie (2007), “if nominal income is growing, the appropriate equilibrium condition calls not for a balanced budget but for a deficit big enough to keep the debt growing in proportion to income, the proportion being determined by portfolio considerations”). It makes more sense to think of the deficit as the instrument and maybe full-employment the target, or, including the JG, the deficit and the JG as the instruments and full-employment and price level stability as the targets or policy objectives.
No idea how or why you think the Lucas Critique and Goodhart’s Law are “quite different”, though. Their equivalence is a proposition of such obviousness that even Wikipedia is alive to it. Think of the most famous application of Goodhart’s Law that “any observed statistical regularity will tend to collapse once pressure is placed upon it for control purposes”— the breakdown of the statistical relationship between inflation and unemployment under pressure from policy makers. Oh, wait a sec, I mean the most famous application of the Lucas Critique—don’t I…?
February 3rd, 2011 at 7:32 am PST
link
Hi Vimothy,
Lucas critique is about regime change and behavior change. Goodhart’s law is about the difference between correlation and causation, not necessarily behavioral change in response to policy change. And, actually, the most famous application of Goodhart’s law is targeting the money supply as a means to macro stabilization–again, confusing correlation with causation. Not surprising that Wiki gets it wrong. They can appear similar, but technically they’re not the same (though perhaps not “quite different” as I wrote above). Goodhart’s smart enough and well versed enough in the field that he would have known he was just redressing the Lucas critique if that’s all he was intending to do.
I would completely agree that targeting NFA would be very suspect to Goodhart’s law, btw.
Best,
Scott
February 3rd, 2011 at 3:52 pm PST
link
Hi Scott,
Yes, the most famous application of Goodhart’s Law is macro stabilization policy targeting the money supply. I was being somewhat facetious above. Of course the most famous application of the Lucas Critique—not Goodhart’s Law–is the breakdown of the correlation between unemployment and inflation. You can understand the jist of the Lucas Critique in very general terms—it doesn’t have to be about regime change, you can think of it as being careful in what variables you hold constant. Specifically, Lucas asks if your parameters are policy invariant (regime invariant)—which is exactly what Goodhart’s Law asks. Goodhart’s Law is not simply an admonition not to confuse correlation with causation. It really would be redundant in that case.
My favorite example of the Lucas Critique is UK education policy under New Labour (in an earlier life I worked for various education research projects)—School Effectiveness / School Improvement turned largely how one might expect.
Incidentally, Goodhart’s Law actually slightly predates the Lucas Critique: “Monetary Relationships: A View from Threadneedle Street” was published in 1975, whereas “Econometric Policy Evaluation: A Critique” was published in 1976. Great minds think alike.
February 4th, 2011 at 10:51 am PST
link
Scott,
Apologies for that rather garbled comment–I was in a rush when I wrote it. Hopefully, you can still make out my position through the poorly constructed sentences.
This does seem like a slightly trivial issue to be debating, though. Can we discuss something more substantive, perhaps? I’m intrigued by your concession re NFA targeting. What do you see as the downside risks to such a policy–losing control of inflation?
February 4th, 2011 at 2:05 pm PST
link
Hi Vimothy,
Good catch on LC predating GL. Got me there. I do think they are different, but, yes, trivial overall either way.
Regarding targeting NFA, I don’t know how you could really do it. The appropriate NFA would seem to be such a complete moving target. I don’t know that it’s a concession. We look at NFA as an indicator (particularly during an asset price bubble period like 1995-2008, and then the balance recession that followed, but also for the state of the economy), but I can’t say I’ve seen anyone suggest it as an actual intermediate policy target. Perhaps an MMT’er has and I missed it. By way of analogy, Goodhart sees the money supply as an indicator for policy makers to use, but obviously understands it can’t be an intermediate target.
That’s why we’ve always emphasized functional finance–defined as being concerned with the effects of deficits (NFA for our purposes here), not the size. So, it’s clearly full employment without accelerating inflation that we are interested in, and not ever NFA as more than just one indicator for the macro economy and Minskian fragility (although sometimes a very important one).
Functional finance is the rationale for something like the JG–if (and it’s admittedly a very big “if,” politically, administratively, and so forth) it could get up and running such that it actually did rise and fall counter to private sector hiring/laying off, it could be a strong macro stabilizer, and an automatic one at that. This essentially would target total employment, again, as the JG “pool” varied opposite to the private sector’s job creation (lots more could be said there, but leave it at that for now).
The JG may not be enough sometimes, we acknowledge, like in balance sheet recessions, or the creation of asset price bubbles that ultimately lead to balance sheet recessions, for which in our view a more Minskyan approach to financial regulation is also required. And there are other ideas for complementing the JG even in more traditional countercyclical policy (i.e., not balance sheet recessions or asset price bubbles) that would largely be additional automatic stabilizers.
Also, note that we aren’t interested in a macro policy that stabilizes the macro economy “perfectly” such that the business cycle is eliminated–as Mosler likes to say, “capitalism is pro-cyclical, get over it.” Some (though nobody in particular comes to mind at the moment) might think we are from some of our blogs, but it actually goes against the very fundamentals of the Minskyan strand of MMT.
February 5th, 2011 at 5:28 pm PST
link
Monetary flows (the proxy for real-output), is going to approach, if not set, millenia records, in the 1st & 2nd qtrs of 2011.
Greenspan, just like Volcker, NEVER tightened monetary policy (despite 17 raises in the FFR over 41 consecutive months). However Bernanke tightened monetary policy for 29 consecutive months, NEVER easing when Bear Sterns 2 hedge funds collapsed, but waiting until Lehman Bros. declared bankruptcy. I.e., Bernanke continued to drive the economy deeper into our Great Recession even while the economy was contracting.
Buying governments isn’t going to reduce unemployment, because, if there is an inflation-unemployment trade-off curve, it is shifting to the right, & at an accelerated rate.
Bernanke’s principle reliance on the Commercial and the Reserve Banks to increase aggregate monetary purchasing power is doomed to fail. Only the loans & investments made by the financial intermediaries (non-banks & shadow banks) match savings with investment. CBs always create new money when they lend & invest. They do not loan out existing deposits, saved or otherwise. I.e., lending by the CBs is inflationary. But lending by the non-banks is not. The correct paradigm is 1966.
February 5th, 2011 at 9:37 pm PST
link