tag:blogger.com,1999:blog-2499715909956774229.post4222920190195808973..comments2024-07-06T08:44:20.403-07:00Comments on Stephen Williamson: New Monetarist Economics: Neo-Fisherian DenialStephen Williamsonhttp://www.blogger.com/profile/01434465858419028592noreply@blogger.comBlogger36125tag:blogger.com,1999:blog-2499715909956774229.post-15566229273395919492016-05-29T09:19:23.017-07:002016-05-29T09:19:23.017-07:00"the inflation rate decline preceded the inte..."the inflation rate decline preceded the interest rate decline."<br /><br />There are many factors that affect the inflation rate, in addition to monetary factors that are under the control of the central bank.Stephen Williamsonhttps://www.blogger.com/profile/01434465858419028592noreply@blogger.comtag:blogger.com,1999:blog-2499715909956774229.post-80014199045489079982016-05-28T23:56:41.011-07:002016-05-28T23:56:41.011-07:00But clearly, Prof, the inflation rate decline prec...But clearly, Prof, the inflation rate decline preceded the interest rate decline. Dr Richard Werner has clearly said something like this: <br /><br />[Lower interest rates do not drive the real economy, but rather are a result of low growth. So, to say low rates lead to high growth and high rates leads to low growth is wrong. Correctly speaking, low growth leads to low rates and high growth leads to high rates.]<br /><br />I am not against raising interest rates some, because it may encourage banks to lend, if there is a little growth already. We don't want to be Europe where people are protesting against the Rothschild ECB. Gary Andersonhttp://www.talkmarkets.com/content/us-markets/will-fed-and-central-bankers-give-up-alchemy-to-save-the-world?post=74282&uid=4798noreply@blogger.comtag:blogger.com,1999:blog-2499715909956774229.post-27658551208439040152016-05-28T12:18:48.008-07:002016-05-28T12:18:48.008-07:00What he finds "persuasive" is convention...What he finds "persuasive" is conventional wisdom, which is not helping anyone, including the majority of the world's central bankers, understand what is going on.Stephen Williamsonhttps://www.blogger.com/profile/01434465858419028592noreply@blogger.comtag:blogger.com,1999:blog-2499715909956774229.post-2892340388297052832016-05-28T11:59:26.101-07:002016-05-28T11:59:26.101-07:00Dr Williamson, the blog at WSJ takes issue with Ne...Dr Williamson, the blog at WSJ takes issue with NeoFisherian thought: <br /><br />"For neo-Fisherism to be right, the private sector would respond to higher interest rates by both spending less and expecting higher prices. For example, firms would respond to falling sales by raising prices. This doesn’t seem to be how firms and workers usually behave (there are exceptions). If it was, it would make central bankers’ jobs a nightmare."<br /><br />and:<br /><br />"It is far more persuasive that low inflation leads to lower interest rates and the reason inflation hasn’t risen is that rates should have been even lower. Risk aversion, tepid investment and fiscal austerity mean demand has been less responsive to a given interest rate. In fact, the equilibrium real rate may be negative. The Fed can’t push nominal rates below zero, so monetary policy has been too tight."<br /><br />So, clearly, at the time of the greatest crash, say mid 2008, inflation started to decline and lower rates were a response to that interest rate decline, not the other way around. <br /><br />Could you respond? Here is the blog link.<br /><br />http://blogs.wsj.com/economics/2015/12/08/greg-ip-could-higher-interest-rates-lead-to-higher-inflation-explaining-neo-fisherism/Gary Andersonhttp://www.talkmarkets.com/contributor/Gary-Anderson/content-article?uid=4798noreply@blogger.comtag:blogger.com,1999:blog-2499715909956774229.post-28069346035955265072016-05-13T06:04:09.123-07:002016-05-13T06:04:09.123-07:00"...the mandate is to achieve price stability..."...the mandate is to achieve price stability and maximum employment. No interpretation there."<br /><br />Of course there is. Price stability is vague enough, but maximum employment is even more so. You can interpret that in many ways.<br /><br />"...in many other countries the inflation target is assigned by the government..."<br /><br />Not quite correct. Usually there's an agreement between government and central bank - e.g. Canada or New Zealand.<br /><br />"The non-linear NK has a continuum of steady states indexed by the inflation rate."<br /><br />Wrong.<br /><br />"there's nothing that says that the interest rate pins down inflation."<br /><br />There is something. It's the Euler equation ("IS curve") that pins it down.<br /><br />"how many more years of zero interest rates and inflation expectations at 2%"<br /><br />By some measures, inflation expectations are well below 2%. And for those who still think that inflation will be 2% in the future, keep watching actual inflation, and adapt.<br /><br />"Neo-Fisherian interpretation"<br /><br />Not an interpretation. These are the implications of the model.<br /><br />Stephen Williamsonhttps://www.blogger.com/profile/01434465858419028592noreply@blogger.comtag:blogger.com,1999:blog-2499715909956774229.post-41888014036058110252016-05-13T03:30:57.867-07:002016-05-13T03:30:57.867-07:00You're right that the Fed has interpreted 2% a...You're right that the Fed has interpreted 2% as the inflation target, but the mandate is to achieve price stability and maximum employment. No interpretation there. And in many other countries the inflation target is assigned by the government, without any interpretation from the central bank.<br /><br />I continue to disagree that (6) and (7) say that the nominal interest rate determines inflation expectations. The non-linear NK has a continuum of steady states indexed by the inflation rate. You can show that zero inflation is optimal in certain circumstances, but there's nothing that says that the interest rate pins down inflation. In fact, you can solve for the optimal inflation rate without any reference to the interest rate!<br /><br />As for the ZLB, how many more years of zero interest rates and inflation expectations at 2% do you need to be convinced that the Fed is fully credible? That's exactly the reason why it can stay at the ZLB<br /><br />Maybe you are searching for a better model of credibility, and you may not like the assumption of the NK model. Fine. But I think the Neo-Fisherian interpretation you are spinning on the model is just wrong.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-2499715909956774229.post-67289911328400563712016-04-28T08:54:33.415-07:002016-04-28T08:54:33.415-07:00Yes, in a steady state, output and inflation are c...Yes, in a steady state, output and inflation are constant. There can be plenty of non-steady state equilibria, though. Under some policy rules, those non-steady steady equilibria will converge to a steady state. But there are some cases, for example in the Taylor rule perils chart, with d > 1, where there are equilibria for which inflation grows without bound. The simple case supposes that (2) always holds, which has output exploding. That's not very attractive, so you can modify it. Indeed, you can get the same dynamics if you just assume y(t) fixed (cash-in-advance endowment economy for example), instead of (2). That doesn't violate any boundary condition. You can get these types of hyperinflationary equilibria in simple monetary models. For example, overlapping generations models of money can have many equilibria that converge in the limit to valueless money.Stephen Williamsonhttps://www.blogger.com/profile/01434465858419028592noreply@blogger.comtag:blogger.com,1999:blog-2499715909956774229.post-57315940518378071252016-04-27T14:29:27.433-07:002016-04-27T14:29:27.433-07:00Hmm, on the first point, it is because a steady-st...Hmm, on the first point, it is because a steady-state is defined as pi(t) = pi(t-1), right? But why need that be the case? I understood that it is because otherwise you end up with ever increasing or decreasing inflation, which violates some boundary condition. Or is there some other reason? Anyway, if the results are robust then that certainly strengthens the argument. But I need to think more about this. Constantine Alexandrakishttps://www.blogger.com/profile/03148709241309623293noreply@blogger.comtag:blogger.com,1999:blog-2499715909956774229.post-87265658623533148982016-04-27T12:08:01.652-07:002016-04-27T12:08:01.652-07:001. (6) is just a property of a steady state. Doesn...1. (6) is just a property of a steady state. Doesn't say anything about the set of equilibria.<br />2. It's correct that to get (1), you need a condition that prevents ponzi schemes on the part of the agents in the model.<br />3. On infinite output, if you impose a condition that bounds labor supply, so that output must be finite, I think you can still have inflation running off to infinity in equilibrium. You have to alter (2), though, to account for that.<br /><br />None of that alters what I discuss above. Basically, the standard policy rules people talk about have bad properties, but neo-Fisherian rules have good properties, and there are many ways in which neo-Fisherian rules give the economy Fisherian properties.Stephen Williamsonhttps://www.blogger.com/profile/01434465858419028592noreply@blogger.comtag:blogger.com,1999:blog-2499715909956774229.post-44421029035746992802016-04-27T11:50:06.066-07:002016-04-27T11:50:06.066-07:00My concern is that some of the results seem to be ...My concern is that some of the results seem to be driven by transversality conditions that provide the additional equations necessary to close the system. For example, that's how you get (6) from (1) (by imposing non-explosive inflation). Similarly, equilibria are ignored if they lead to infinite output in finite time. Now, if economic agents can tell when such a condition is violated and realize that this cannot be an equilibrium then this common assumption would be harmless. But I don't think this happens in real life. Usually there is some mechanism that kicks in if the economy is an unsustainable trajectory, and if it takes a long time until that mechanism kicks in then people may not be aware that some equilibria have bizarre properties, in which case policy could work in ways very different from what the models predict. Constantine Alexandrakishttps://www.blogger.com/profile/03148709241309623293noreply@blogger.comtag:blogger.com,1999:blog-2499715909956774229.post-45310224139507834372016-04-23T12:47:46.074-07:002016-04-23T12:47:46.074-07:00"you basic misunderstanding is that you have ..."you basic misunderstanding is that you have the steady state implication of (6) wrong..."<br /><br />I'd say you have it wrong. Congress gives the Fed some vague instructions about what it is supposed to do, and the Fed has interpreted that as a 2% inflation target and "maximum employment." In any case, it's not the inflation target that determines long-run inflation expectations, it's the long run nominal interest rate, in this model. That's what (6) and (7) say. But, given an inflation target, the question is: given this model, how do you deliver it? Addendum #3 is in fact an optimal rule, I'm convinced. Given this rule, the central bank hits its inflation target in every period but the first (as in this model, the central bank is essentially stuck with first period inflation). Inflation expectations are indeed then anchored at the inflation target forever. But, if the central bank follows a conventional Taylor-principle rule, expectations become "unanchored" in the sense that, for example, the nominal interest rate can go to the ZLB and stay there. When inflation expectations become anchored again, they are anchored at a low anticipated inflation rate. If the central bank is claiming that it can achieve its inflation target by staying at the ZLB, then it's lost all credibility.<br /><br />"By the way, using the linearized version of the model as if those were levels is also misleading."<br /><br />In the deterministic version of the model, (1) is literally what you get when you take logs on either side of the Euler equation. So, it's not an approximation. The Phillips curve is whatever it is. Nothing misleading at all.<br /><br />Stephen Williamsonhttps://www.blogger.com/profile/01434465858419028592noreply@blogger.comtag:blogger.com,1999:blog-2499715909956774229.post-22255365231878705242016-04-21T02:11:29.026-07:002016-04-21T02:11:29.026-07:00I think you are at stage 1 of understanding the NK...I think you are at stage 1 of understanding the NK model. Anchored inflation expectations doesn't mean exogenous. In fact, you can solve the NK model with a forward looking Phillips curve paper and pencil even without abstracting from expected inflation. Importantly, you can solve for expected inflation. You will see that expected inflation does move. Anchored inflation expectations means that IN THE LONG RUN inflation expectations converge to the inflation target. <br /><br />More constructively, you basic misunderstanding is that you have the steady state implication of (6) wrong. Congress gives the Fed an inflation target. What anchors inflation expectations is the credibility of the Fed, and the coordination of beliefs around that target. Equation (3) determines the nominal interest rate that supports that inflation rate. For example, if the target is zero, the nominal rate has to be equal to the real rate, which is just the inverse of the individual discount factor.<br /><br />By the way, using the linearized version of the model as if those were levels is also misleading.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-2499715909956774229.post-38750929738551239232016-04-20T14:22:55.377-07:002016-04-20T14:22:55.377-07:00negative rate proposals usually apply just to the ...negative rate proposals usually apply just to the short end of the yield curve - typically it's the interest rate on reserves that goes negative. Janet Yellen was asked a question about negative interest on reserves, either at the last press conference, or when she talked to Congress. You should be able to find that on video.Stephen Williamsonhttps://www.blogger.com/profile/01434465858419028592noreply@blogger.comtag:blogger.com,1999:blog-2499715909956774229.post-19787390575564887882016-04-18T18:27:49.196-07:002016-04-18T18:27:49.196-07:00Dr Williamson, interesting comment. So, to avoid ...Dr Williamson, interesting comment. So, to avoid potential negative nominal bond yields getting out of hand in a recession, would the Fed ever be interested in doing the Market Monetarist thing of negative IOR to spur lending? I am assuming that negative IOR can be accomplished without pushing bonds into negative territory. <br /><br />And, whether Phelan is right or not, I know the Fed does seem to mandate slow growth in other ways. Would you agree with that? Thanks for your time, btw. Gary Andersonhttp://www.talkmarkets.com/content/economics--politics-education/federal-reserve-mandates-slow-growth-so-fed-must-finance-american-infrastructure?post=91756&uid=4798noreply@blogger.comtag:blogger.com,1999:blog-2499715909956774229.post-91712013789426573722016-04-18T17:36:43.798-07:002016-04-18T17:36:43.798-07:00Saw the Phelan paper, and didn't find it compe...Saw the Phelan paper, and didn't find it compelling.Stephen Williamsonhttps://www.blogger.com/profile/01434465858419028592noreply@blogger.comtag:blogger.com,1999:blog-2499715909956774229.post-69773565651045067312016-04-18T09:38:57.648-07:002016-04-18T09:38:57.648-07:00Well, the GDP was being destroyed in 2007/2008. No...Well, the GDP was being destroyed in 2007/2008. Not sure monetary stimulus could not have helped in non bubble areas where employment was destroyed even where there were no house bubble bursts. <br /><br />Also, question, does the Fed agree with Chris Phelan that there could be a run on the Fed if all the banks decided to lend out the excess reserves? Seems like a little lending would be better than negative interest rates if it comes to that. Gary Andersonhttp://www.talkmarkets.com/content/us-markets/federal-reserve-mandates-slow-growth-so-fed-must-finance-american-infrastructure?post=91756noreply@blogger.comtag:blogger.com,1999:blog-2499715909956774229.post-86223499591119257672016-04-18T06:43:45.221-07:002016-04-18T06:43:45.221-07:00The standard intuition, I think, is that expectati...The standard intuition, I think, is that expectations are "anchored," which I take to mean exogenous. So, if the current nominal rate goes up, the real rate goes down. Further, standard intuition is that future consumption and output are exogenous in this experiment. So, if the real interest rate is higher, standard consumption-smoothing dictates that current consumption should fall. Current output is lower, and then a Phillips curve effect dictates that current inflation is lower (given anchored expectations). That's basically what's going on in (3), if you fix the second term on the right-hand side.<br /><br />For some reason, many people seemed convinced that this is what is going on in the NK model. But that's not correct. When R is raised permanently, inflation expectations are not fixed - they have to adapt to the policy. The real rate is higher alright, and that's reflected in lower consumption today than in the future, but that need not mean that current consumption goes down. It actually means that future consumption goes up. And that higher future consumption and output, working through the Phillips curve effect, is pushing up inflation over time. That's how it works in this model, anyway.Stephen Williamsonhttps://www.blogger.com/profile/01434465858419028592noreply@blogger.comtag:blogger.com,1999:blog-2499715909956774229.post-21674046429083323822016-04-17T13:35:45.074-07:002016-04-17T13:35:45.074-07:00Well, my intuition is the following. The cb perman...Well, my intuition is the following. The cb permanently increases the nominal rate. People want to save more and spend less on goods. This should induce producers to lower prices. However, people save more and spend less only if higher nominal rate is not compensated by higher expected inflation, so a sufgiciently higher future price level can avoid the fall in today's prices, and the latter is the neo fisherian case. What I dont get is what should induce producers to raise prices over time. I am completely at a loss on this. The only idea I can have is that this happens through some form of self fullfilling effect, like people perfectly believes in the cb inflation target and so raise prices over time. If this is the reason, it is quite unintuitive.LMhttps://www.blogger.com/profile/04209992315578358377noreply@blogger.comtag:blogger.com,1999:blog-2499715909956774229.post-22493057512532558352016-04-17T07:43:59.488-07:002016-04-17T07:43:59.488-07:00Once we include fiscal policy and start thinking a...Once we include fiscal policy and start thinking about the fiscal/monetary interaction, lots of things can happen. Sufficiently rich tax policy and the zero lower bound is irrelevant, for example, in NK models.Stephen Williamsonhttps://www.blogger.com/profile/01434465858419028592noreply@blogger.comtag:blogger.com,1999:blog-2499715909956774229.post-52418183728313725332016-04-17T07:41:46.721-07:002016-04-17T07:41:46.721-07:00If we have a cashless model in which prices are fu...If we have a cashless model in which prices are fully flexible, then we're just adding an irrelevant numeraire to the model. Then "monetary policy" is also irrelevant. But, you could do simple monetary models with no sticky prices, and work out inflation dynamics under different Taylor rules. Roughly, the properties will be the same as what I worked out here.Stephen Williamsonhttps://www.blogger.com/profile/01434465858419028592noreply@blogger.comtag:blogger.com,1999:blog-2499715909956774229.post-22817390251039778732016-04-17T07:37:23.048-07:002016-04-17T07:37:23.048-07:00Not being intuitive just means you don't get i...Not being intuitive just means you don't get it. So, tell me how you think monetary policy works, and where your intuition about that comes from.Stephen Williamsonhttps://www.blogger.com/profile/01434465858419028592noreply@blogger.comtag:blogger.com,1999:blog-2499715909956774229.post-85062683637479323072016-04-16T20:22:08.642-07:002016-04-16T20:22:08.642-07:00What about models in which the fiscal authority re...What about models in which the fiscal authority responds to higher interest rates by reducing its deficits over the medium and long term?<br /><br />Couldn't this result in a model where:<br />1) "tight money" causes a short-term contraction due to frictions such as sticky prices.<br />2) The threat of higher interest rates induces a fiscal contraction over the medium/long term.<br />3) Expectations of lower inflation due to the fiscal contraction cause the rate of inflation to go back down again, which is then accommodated by the central bank lowering its interest rate again.Dimitrihttps://www.blogger.com/profile/00802842224248550248noreply@blogger.comtag:blogger.com,1999:blog-2499715909956774229.post-1335088481720667302016-04-16T10:20:29.375-07:002016-04-16T10:20:29.375-07:00But none of the models shown by stephen has any of...But none of the models shown by stephen has any of these features, which make the neo fisherian result quite unintuitive. I understand the neo fisherian result only in a ftpl framework, where higher rates reduce ability of government to pay back debt and people get rid of government debt to buy things. Without that I still find it hard to understand neo fisherism, not saying it s wrong, just unintuitive.LMhttps://www.blogger.com/profile/04209992315578358377noreply@blogger.comtag:blogger.com,1999:blog-2499715909956774229.post-65713108001589954412016-04-16T05:31:23.529-07:002016-04-16T05:31:23.529-07:00Another thing that's still not completely clea...Another thing that's still not completely clear, or maybe I'm just lazy to fully do the derivations on Saturday afternoon: consider the flexible price model underlying the New Keynesian economy and assume a permanent rise in the nominal interest rate. The Fisher equation says we need to have expected inflation between now and next period. But how do we know if this comes through a rise in the price level relative to the previous level, or rather through a quick decline in prices in the current period, followed by a rebound to the same price level next period. Something in the nominal equilibrium selection, since a cashless/money as pure numeraire flex price economy only determines expected inflation, not the price level? I think this matters in getting the results for example in Cochrane's analysis, where inflation rises even with sticky prices. If we selected an equilibrium where instead with flexible prices prices first drop (i.e you have deflation relative to the initial period), providing room for expected inflation next period, then a local to friction-less sticky price equilibrium with rational expectations should also have short-term lower inflation followed by higher inflation. What selects the price level here (similar questions arise in open economy models where the UIRP holds: you often get an initial exchange rate depreciation that allows for a future appreciation in response to an interest rate cut. If you prolong the initial depreciation a bit due to short term deviation from rational expectations, you get a carry trade).danielshttps://www.blogger.com/profile/01799942447501959179noreply@blogger.comtag:blogger.com,1999:blog-2499715909956774229.post-53250427774556007392016-04-15T11:13:16.169-07:002016-04-15T11:13:16.169-07:00In that case, though, the financial crisis was dri...In that case, though, the financial crisis was driving a drop in asset prices, including the prices of commodities, so I wouldn't attribute much of that to monetary factors.Stephen Williamsonhttps://www.blogger.com/profile/01434465858419028592noreply@blogger.com