tag:blogger.com,1999:blog-8702840202604739302.post7617203075027143462..comments2024-03-12T22:00:25.991-07:00Comments on MacroMania: A monetary-fiscal theory of inflationDavid Andolfattohttp://www.blogger.com/profile/12138572028306561024noreply@blogger.comBlogger27125tag:blogger.com,1999:blog-8702840202604739302.post-57463278842908983212017-08-30T11:23:09.777-07:002017-08-30T11:23:09.777-07:00Oh, it's equation 2. The usual mechanism, sil...Oh, it's equation 2. The usual mechanism, silly question.Adam Phttps://www.blogger.com/profile/16316584837610367439noreply@blogger.comtag:blogger.com,1999:blog-8702840202604739302.post-51479980713582081972017-08-30T09:05:34.224-07:002017-08-30T09:05:34.224-07:00That's from the CIA constraint on investment, ...That's from the CIA constraint on investment, your analogue of a reserve requirement?Adam Phttps://www.blogger.com/profile/16316584837610367439noreply@blogger.comtag:blogger.com,1999:blog-8702840202604739302.post-25073412511811126762017-08-30T08:55:58.094-07:002017-08-30T08:55:58.094-07:00As long as the yield on treasuries exceeds the yie...As long as the yield on treasuries exceeds the yield on reserves, then in this model QE has the standard textbook effects. An increase in money/bond ratio will increase the price-level, lower the real interest rate, and stimulate economic activity. If the yield on treasuries and reserves is identical, then the effect of QE is inconsequential. <br /><br />See: <br /><br />https://files.stlouisfed.org/files/htdocs/publications/review/2015-09-08//a-model-of-u-s-monetary-policy-before-and-after-the-great-recession.pdfDavid Andolfattohttps://www.blogger.com/profile/12138572028306561024noreply@blogger.comtag:blogger.com,1999:blog-8702840202604739302.post-21567288693248166822017-08-30T04:43:49.745-07:002017-08-30T04:43:49.745-07:00David,
In this model what, if anything, does QE a...David,<br /><br />In this model what, if anything, does QE accomplish?Adam Phttps://www.blogger.com/profile/16316584837610367439noreply@blogger.comtag:blogger.com,1999:blog-8702840202604739302.post-51104042044517628672017-08-25T02:50:01.491-07:002017-08-25T02:50:01.491-07:00I think we're communicating now, we agree on t...I think we're communicating now, we agree on the whole story I think with one exception: "forget about the UR".<br /><br />I don't think that's a good idea, even accepting the rest of the theoretical argument, and the fact that you can "write down a model..." is not a good reason.<br /><br />I can write down a model where UR causes inflation, but the reason we're having this conversation is our agreement that such a model won't be a satisfactory description of reality.<br /><br />My insistence that the UR is a valid indicator of is both empirical and theoretical. Empirically, as you pointed out in your last post, the downward sloping PC relationship does seem to be there an awful lot of the time. I take that to indicate that in the absence of other factors which must be somewhat unusual, shifts in the relative supply/demand for money and government debt effects both variables (shifts that raise inflation usually lower UE).<br /><br />I suspect that your wish to forget about UR is not so much you disagree with the last sentence of that paragraph I just wrote but that you think it's all subsumed by observing inflation. Once you see the price level the UR adds no more information. I claim it does, take the most recent fall in inflation, apparently due at least in large part to cell phone pricing and prescription drugs if I understand correctly. Do we think that was a sudden increase in demand for money and bonds? I think it was a some large relative price changes that just happened to coincide in timing and direction, it was a random fluctuation in the measured inflation rate. If so, perhaps it should be ignored by the fed, just as some random upwards fluctuations were ignored (apparently correctly). The main thing that makes me believe this is the continued falling UR, which shouldn't be happening if there was a fundamental shift in the supply/demand for money and government debt.Adam Phttps://www.blogger.com/profile/16316584837610367439noreply@blogger.comtag:blogger.com,1999:blog-8702840202604739302.post-2983671713129948532017-08-24T16:06:08.086-07:002017-08-24T16:06:08.086-07:00Hmm, maybe I don't understand what you are say...Hmm, maybe I don't understand what you are saying. <br />Let's see where I think we agree.<br /><br />1. Forget PC theory of inflation. Check. What this means is that unemployment has no direct causal effect on inflation. Right?<br /><br />2. Focus on supply/demand for money/debt. Check. Cannot directly observe demand. Check. Look to low inflation to diagnose high money demand (relative to supply, which we can measure). Check. <br /><br />The next part is where I lose you. You want another diagnostic for unobservable money demand. You choose unemployment. You claim that low unemployment indicates fall in money demand. This could be the case, but how do you know? I can write down a model where recruiting expense must be financed largely by cash. In this world, a boom in recruiting activity increases the demand for cash, lowering the inflation rate. At the same time, unemployment falls. How does low UR in this model portend higher future inflation? It does not--it is the opposite in this case.<br /><br />So my basic point is: forget about the UR. If inflation and nominal yields are low money and debt supply is growing, this is enough to tell us what's happening to the demand for money/debt. I don't need to refer to the UR (which could go either way, theoretically).<br /><br />I'm sorry if I once again missed your point. I'm happy to spend the time to try to work it out though! Thx. :)David Andolfattohttps://www.blogger.com/profile/12138572028306561024noreply@blogger.comtag:blogger.com,1999:blog-8702840202604739302.post-33096745755733056532017-08-24T14:22:12.469-07:002017-08-24T14:22:12.469-07:00Do you ever sit down and say, "You know what,...Do you ever sit down and say, "You know what, I'm pretty much a kook?"Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-8702840202604739302.post-29570505203617115752017-08-24T13:56:17.585-07:002017-08-24T13:56:17.585-07:00" forget about the Phillips curve theory of i..." forget about the Phillips curve theory of inflation...Better to focus on the supply and demand for money and debt."<br /><br />For crying out loud David that's what I was doing! Why are you not agreeing with me?<br /><br />But of course, how exactly do I focus on the supply and demand for money and debt? Do I directly observe these quantities? Maybe you do, I don't.<br /><br />So what can I look at that will indicate shifts in the relative supply and demand for money and debt? The first thing I think of is inflation, it's low so ok. But, as you correctly state there are countervailing forces that mess with any theoretical relationship, so I should also look for other indicators.<br /><br />Well, the second one I can think of is unemployment. If it is low and falling then that also says something about the relative supply and demand for money and debt. This is the content of the final paragraph of my original comment, low and falling UE indicates a relative fall in demand for money and government debt and that in turn indicates higher inflation in the future.<br /><br />I don't think I'm taking the formal model too literally, I think I'm taking your theoretical argument seriously.Adam Phttps://www.blogger.com/profile/16316584837610367439noreply@blogger.comtag:blogger.com,1999:blog-8702840202604739302.post-39973508893985304812017-08-24T06:14:52.015-07:002017-08-24T06:14:52.015-07:00I understand what you are saying now. I think you ...I understand what you are saying now. I think you are taking the model too literally. Yes, in the manner it is formulated, what you say is true: ceteris paribus, the model implies low U should forecast high P. <br /><br />However, it would be easy to append to the model a countervailing force where low U is associated with low P. <br /><br />Such a model would contain my original message: forget about the Phillips curve theory of inflation. The direction of causality runs the other way (whether it's negative or positive depends on countervailing forces). Better to focus on the supply and demand for money and debt.David Andolfattohttps://www.blogger.com/profile/12138572028306561024noreply@blogger.comtag:blogger.com,1999:blog-8702840202604739302.post-49742223733485060672017-08-24T03:12:22.218-07:002017-08-24T03:12:22.218-07:00Yes, I understand that in the last post you simply...Yes, I understand that in the last post you simply wanted to give en example of causality running the other way. It's the reference to the Tobin effect that connects to the model you are describing here, you use the Tobin effect first as an example of a downward sloping PC with the P to U causation and then the same effect to get the PC in this model. Hence I conclude the model being discussed here has a downward sloping PC.<br /><br />Which brings me to my point, that you have yet to address. In this post you assert that if U does not cause P, still in the context of a downward sloping PC, then low U is not a reason to raise rates. However, if your downsloping PC is an equilibrium condition then we should expect the economy to be on the curve. If so then the direction of causality is irrelevant, observing low U tells us to expect higher P even if causation runs from P to U a la Tobin. The data is very noisy and so other indicators besides attempts at direct measurements are useful for understanding inflation. Your model says U is such an indicator.<br /><br />Now, my last paragraph was really about reconciling all this with the fact that while U seems low, inflation is not high. The point here is that in your model low and falling U indicates less demand (relative to supply) for government debt and your model would say this should lead to higher inflation. This, it seems to me that your model does in fact say that low U is a valid reason to tighten in expectation of higher inflation.<br />Adam Phttps://www.blogger.com/profile/16316584837610367439noreply@blogger.comtag:blogger.com,1999:blog-8702840202604739302.post-70113642317927203262017-08-23T13:17:50.318-07:002017-08-23T13:17:50.318-07:00I had in mind that it's possible to have a mod...I had in mind that it's possible to have a model where inflation and unemployment were negatively correlated, but where the direction of causality was reversed. The "existence" I referred to was the statistical relation, not the alleged causal relation of "U causes P." <br /><br />Hope this clears things up. David Andolfattohttps://www.blogger.com/profile/12138572028306561024noreply@blogger.comtag:blogger.com,1999:blog-8702840202604739302.post-830853537891433082017-08-21T11:24:36.167-07:002017-08-21T11:24:36.167-07:00I didn't think the downward sloping PC was my ...I didn't think the downward sloping PC was my assumption, I thought it was yours!<br /><br />Here you are in your last post, which you explicitly related to this post:<br /><br />"In fact, economists have known for a long time that there are many other mechanisms that might generate a negative relationship between the unemployment rate and inflation. The Tobin effect, for example, asserts that the direction of causality runs in the opposite direction: higher inflation induces a portfolio substitution out of government securities into private investment (including recruiting investment), which leads to lower unemployment. "<br /><br />And here you are in this post:<br /><br />"This is all based on standard Monetarist thinking--we do not need the Phillips curve (which, by the way, exists in my model via a Tobin effect)."<br /><br />Put those two together and it sure sounds you had in mind lower unemployment going with higher inflation as an equilibrium condition.Adam Phttps://www.blogger.com/profile/16316584837610367439noreply@blogger.comtag:blogger.com,1999:blog-8702840202604739302.post-59565714138110367002017-08-21T06:46:55.664-07:002017-08-21T06:46:55.664-07:00I'm struggling with your (implicit) contention...I'm struggling with your (implicit) contention that the unemployment rate should be an indicator of higher inflation. Theoretically, we can get the PC to slope any which way we want. To the extent that inflation is a tax, higher inflation causes higher unemployment, for example. David Andolfattohttps://www.blogger.com/profile/12138572028306561024noreply@blogger.comtag:blogger.com,1999:blog-8702840202604739302.post-59613020165376142902017-08-20T10:59:28.024-07:002017-08-20T10:59:28.024-07:00Hi David,
I liked the note a lot. In some sense ...Hi David,<br /><br />I liked the note a lot. In some sense better than the post because you dwell there on the point that in fighting low inflation the monetary authority doesn't hold any cards, unlike when fighting high inflation. That's a very good point and I'm not sure I've seen it made so clearly before.<br /><br />However, I'm struggling slightly with the contention that low unemployment isn't an indicator of higher inflation. Perhaps not the cause in the usual sense but, as Cochrane likes to say, equilibrium conditions don't specify a causal ordering. Presumably a Philips curve, even driven by a Tobin effect, is still an equilibrium condition right? And if so, in equilibrium we're gonna be on the curve.<br /><br />As for the causal mechanism in that situation, well you're whole point is about demand to hold government debt. It's higher now then it used to be. But in the Tobin effect the operation of increased inflation just lowers demand for government debt increasing demand for private assets and thus increasing investment behaviour. Any fall in the demand to hold government debt works just as well and low unemployment is the signal that somehow that has happened. After all, is there a way to get such low unemployment without the necessary investment (recruiting investment and other types)?<br /><br />thanks.Adam Phttps://www.blogger.com/profile/16316584837610367439noreply@blogger.comtag:blogger.com,1999:blog-8702840202604739302.post-29605527268089321122017-08-17T04:31:54.620-07:002017-08-17T04:31:54.620-07:00David-
An inflation corridor, or inflation band i...David-<br /><br />An inflation corridor, or inflation band is used by the Reserve Bank of Australia, with arguable success. If you must target inflation, then go with an inflation band, and keep it loose, like 2% to 4%. <br /><br />And what role then for QE?<br /><br />Adair Turner says we should ponder money-financed fiscal programs. I agree.<br /><br />As for the regulatory quagmire that is every developed nation, I am always eager and ready to cut regulations.<br /><br />The first thing to cut and the largest structural impediment: Property zoning. <br /><br />I hope to see a series from you on the depravity of property zoning. <br /><br />Benjamin Colehttps://www.blogger.com/profile/14001038338873263877noreply@blogger.comtag:blogger.com,1999:blog-8702840202604739302.post-86680813671805626772017-08-15T04:56:39.298-07:002017-08-15T04:56:39.298-07:00" I suspect quite a bit of output has been gi..." I suspect quite a bit of output has been given up, possibly even enough to bring us back to the 2008 LT and trend, just to squeeze 1% out of inflation."<br /><br />What makes you say that?<br /> <br />The answer to your last question in my view is yes, at least, in the sense of establishing an inflation corridor. There is little else one can (or should) expect from a central bank. Fiscal policy must take more responsibility during a severe downturn (by increasing transfers, cutting taxes, or both). Much of this already happens naturally via automatic stabilizers. As for secular slowdowns, once again, it is the government's responsibility to review the nature of the regulatory environment and ascertain what if any reforms are needed. This is not a job for the monetary authority. David Andolfattohttps://www.blogger.com/profile/12138572028306561024noreply@blogger.comtag:blogger.com,1999:blog-8702840202604739302.post-44924258195367110592017-08-14T22:12:14.491-07:002017-08-14T22:12:14.491-07:00David A-
Hmm. Let's look at this backwards.
...David A-<br /><br />Hmm. Let's look at this backwards. <br /><br />A decrease in the unemployment rate, even to 3%, may have little or no impact on the reported rate of inflation. <br /><br />see http://ngdp-advisers.com/2017/08/14/st-louis-fed-president-says-3-unemployment-raise-inflation-rate-to1-8/<br /><br />When Blanchard wrote the 2016 paper, unemployment was higher. Looks like the unemployment rate (as measured) fell by a few percent, with no effect on reported inflation. For that matter, inflation has hardly budged since 2008.<br /><br />I suspect quite a bit of output has been given up, possibly even enough to bring us back to the 2008 LT and trend, just to squeeze 1% out of inflation. <br /><br />BTW, from 1982 to 2007, the average US GDP real growth rate was 3+% and the average rate of inflation just under 3%. That is recent history in the US, not theory. So perhaps a lot of real output, profits and wages is being given up to squeeze out the last 1% of inflation. <br /><br />Maybe, as a practical if not theoretical matter, we live in a 3% inflation for 3% real growth or 2 for 2 world. <br /><br />If so, is 3% inflation a good policy goal?<br /><br />Egads, I can remember when the WSJ opined Volcker was too tight, that inflation was under 5%. Milton Friedman chastised the Fed for being too tight in 1993, when inflation was more than 3%. <br /><br />Is a peevish fixation on inflation a good monetary policy? <br /><br /><br /><br />Benjamin Cole <br />Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-8702840202604739302.post-34836046894740481052017-08-12T06:11:06.629-07:002017-08-12T06:11:06.629-07:00It adds this: that policymakers should not expect ...It adds this: that policymakers should not expect labor market tightness to generate *inflation* (as opposed to real wage gains) without accommodation from the monetary and fiscal authorities. David Andolfattohttps://www.blogger.com/profile/12138572028306561024noreply@blogger.comtag:blogger.com,1999:blog-8702840202604739302.post-6695724655546756782017-08-11T23:31:02.227-07:002017-08-11T23:31:02.227-07:00I find this post strange. You begin by writing tha...I find this post strange. You begin by writing that your model is Monetarist but then give reasonable arguments that could have been made just as easily by someone using one of the other three models.<br /><br />Your explanation that the increase in demand for bonds coincided so closely with the increase in the monetary base that it caused neither high inflation or deflation is possible but implausible. Yet the Monetarist framework seems to require this.<br /><br />As you have pointed out several times, the unemployment rate is a very imperfect proxy for slack in the labour market and other measures like the employment rate tell a different story. The Phillips Curve Model allows for almost exactly the same conclusions as the Monetarist regarding the situation we are in. So why use the Monetarist Quantity Theory? What does it add?Anonymoushttps://www.blogger.com/profile/07730077084140053327noreply@blogger.comtag:blogger.com,1999:blog-8702840202604739302.post-68995482353769568802017-08-08T08:06:36.739-07:002017-08-08T08:06:36.739-07:00Ahmad, please see my reply to JP above. Thanks. Ahmad, please see my reply to JP above. Thanks. David Andolfattohttps://www.blogger.com/profile/12138572028306561024noreply@blogger.comtag:blogger.com,1999:blog-8702840202604739302.post-89274354282969090172017-08-08T08:06:05.451-07:002017-08-08T08:06:05.451-07:00JP, all I have in mind is the following. When the ...JP, all I have in mind is the following. When the yield on bonds falls to yield on money, they are essentially perfect substitutes. What is relevant for the price level now is the total money supply, not its composition between money and bonds. Hence, an increase in the demand for bonds is like an increase in the demand for money, and is disinflationary. David Andolfattohttps://www.blogger.com/profile/12138572028306561024noreply@blogger.comtag:blogger.com,1999:blog-8702840202604739302.post-58349617635631726352017-08-07T13:15:51.798-07:002017-08-07T13:15:51.798-07:00Inflation: back to basics
Comment on David Andolfa...Inflation: back to basics<br />Comment on David Andolfatto on ‘A monetary-fiscal theory of inflation’<br /><br />David Andolfatto argues from a sophisticated model: “In my formal model, I have a parameter that indexes the growth rate in the demand for real money/bond balances (where money and bonds take the form of USDs and USTs, respectively). In the open-economy version of my model, I have a ‘money demand growth regime’ originating from the foreign sector. In the model, this regime translates into persistent U.S. trade deficits, representing the foreign sector's desire to acquire USD/UST at an elevated pace.”<br /><br />Basically, in this model deflation/inflation is driven by what happens on the UST market. This is in line with the commonplace Quantity Theory which holds that a smaller or broader composite called ‘quantity of money’ determines the price level.<br /><br />Now, it is well-known that the familiar models, which are either from the Walrasian type (= microfoundations) or the Keynesian type (= macrofoundations), are axiomatically false. Because of this, monetary theory has to be based upon entirely new macrofoundations.#1<br /><br />In order to go back to the basics, the pure consumption economy is for a start clearly defined by three macro axioms (Yw=WL, O=RL, C=PX), two conditions (X=O, C=Yw) and two definitions (profit/loss Qm≡C-Yw, saving/dissaving Sm≡Yw-C).#2<br /><br />Money is needed by the business sector to pay the workers who receive the wage income Yw per period. The workers spend C per period. Given the two conditions, the market clearing price is derived for a start as P=W/R. So, the price P is determined by the wage rate W, which has to be fixed as a numéraire, and the productivity R. From this follows the average stock of transaction money as M=kYw, with k determined by the payment pattern. In other words, the quantity of money M is determined by the AUTONOMOUS transactions of the household and business sector and produced out of nothing by the central bank. The economy never runs out of money.<br /><br />The transaction formula reads M = (k/rhoE)*P*O, with the ratio rhoE defined as C/Yw, and this yields the commonplace correlation between quantity of money M and price P, except for the fact that M is the DEPENDENT variable.<br /><br />The market clearing price is given in the general case with the price formula P = (rhoE)*(W/R). An expenditure ratio rhoE greater than 1 indicates credit expansion = dissaving, a ratio rhoE less than 1 indicates credit contraction = saving. In the initial period rhoE = 1, i.e. the household sector’s budget is balanced. The ratio rhoE establishes the link between the product market and the money/capital market.<br /><br />Now we have: deficit spending, i.e. rhoE greater 1, yields a price hike. If deficit spending is repeated period after period the price remains on the elevated level but there is NO inflation. No matter how long the household sector’s debt increases, there is NO accelerated price increase.<br /><br />The price formula makes it clear that inflation only occurs if the wage rate W increases in successive periods faster than productivity R. This can happen at ANY employment level. It is NOT a precondition that employment is close to the capacity limit. This is merely a false interpretation of the Phillips curve.<br /><br />The current deflationary trend is caused by the fact that (worldwide) wages lag behind productivity growth. To turn this trend around it does not matter much what happens on the market for UST, what matters is that goverments/central banks engineer a coordinated world wide increase of the average wage rate.<br /><br />Egmont Kakarot-Handtke<br /><br />#1 First Lecture in New Economic Thinking<br />http://axecorg.blogspot.de/2017/05/first-lecture-in-new-economic-thinking.html<br /><br />#2 For the detailed description see ‘How the intelligent non-economist can refute every economist hands down’<br />http://axecorg.blogspot.de/2015/12/how-intelligent-non-economist-can.htmlAXEC / E.K-Hhttps://www.blogger.com/profile/10402274109039114416noreply@blogger.comtag:blogger.com,1999:blog-8702840202604739302.post-88512767154635585412017-08-07T13:12:09.511-07:002017-08-07T13:12:09.511-07:00My major is economics. But I didn't get this p...My major is economics. But I didn't get this point: <br /><br />"An increase in the demand for bonds in this case must manifest itself in other ways. One way is for the price-level to fall. That is, a market-mechanism for expanding the real supply of nominal bonds is for the price-level to fall".<br /><br />I would be grateful if you could possibly guide me about it.A.Mhttps://www.blogger.com/profile/06156646553465404381noreply@blogger.comtag:blogger.com,1999:blog-8702840202604739302.post-31731025789820040852017-08-07T07:31:35.807-07:002017-08-07T07:31:35.807-07:00David, assume the central bank can credibly commit...David, assume the central bank can credibly commit to a permanent expansion of the monetary base and this expansion is not sterilized via something like interest on reserves. In this case, the yield on monetary base and treasury bills would not be the same (because of expected inflation) and they would cease to be perfect substitutes. <br /><br />This scenario is unlikely in our world, but it is not impossible. Given this possibility, what does this imply about why the demand for treasury securities matter for spending and inflation?David Beckworthhttps://www.blogger.com/profile/04577612979801459194noreply@blogger.comtag:blogger.com,1999:blog-8702840202604739302.post-89869555858485660752017-08-07T06:37:24.659-07:002017-08-07T06:37:24.659-07:00"An increase in the demand for bonds in this ..."An increase in the demand for bonds in this case must manifest itself in other ways. One way is for the price-level to fall. That is, a market-mechanism for expanding the real supply of nominal bonds is for the price-level to fall."<br /><br />Am struggling to understand exactly how this works. Once bond yields are equal to yields on money, the two are substitutes. Subsequent investor demand for bonds is now just as likely to manifest itself as a demand for money as it is for bonds. And that portion that is expressed by purchases of money can push the price level down if it accommodated by the central bank? Maybe you have something else in mind?JP Koninghttps://www.blogger.com/profile/02559687323828006535noreply@blogger.com