tag:blogger.com,1999:blog-8702840202604739302.post7697435206398954651..comments2024-03-28T03:38:53.734-07:00Comments on MacroMania: Fisher without EulerDavid Andolfattohttp://www.blogger.com/profile/12138572028306561024noreply@blogger.comBlogger9125tag:blogger.com,1999:blog-8702840202604739302.post-78896844022510919832015-11-11T08:29:26.075-08:002015-11-11T08:29:26.075-08:00I think one could view V as being determined by fo...I think one could view V as being determined by forward-looking behavior, but that forward-looking is not equivalent to rational expectations.<br /><br />Good to see you reading FRB St. Louis research! :)David Andolfattohttps://www.blogger.com/profile/12138572028306561024noreply@blogger.comtag:blogger.com,1999:blog-8702840202604739302.post-24155319171205407332015-11-11T06:17:45.742-08:002015-11-11T06:17:45.742-08:00I, as a fairly junior student of economics, wonder...I, as a fairly junior student of economics, wonder how the interpretation of the "expectations formation mechanism" of the NF model is affected were the constant velocity assumption implicit in this discussion, and also, from what I can tell, implicit in Erzo's behavioural rule, removed. <br /><br />I note that QTM on its own predicts a decrease in price levels with an increase in the money supply so long as V decreases more than M increases. <br /><br />If we assume V is dynamic, don't such changes occur due to changes in rational expectations? Can we then really say that the NF position does not require rational expectations? <br /><br />Admittedly, I feel like I am missing something fundamental in this discussion but I thought I would ask!<br /><br /><br />I found these interesting:<br /><br />https://research.stlouisfed.org/publications/review/83/12/Velocity_Dec1983.pdf<br /><br />https://www.stlouisfed.org/On-The-Economy/2014/September/What-Does-Money-Velocity-Tell-Us-about-Low-Inflation-in-the-US<br /><br /> Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-8702840202604739302.post-59541745673630806732015-11-11T04:40:25.750-08:002015-11-11T04:40:25.750-08:00[Note: Biago, I accidentally deleted your comment,...[Note: Biago, I accidentally deleted your comment, but recovered it from another source. Here it is.]<br /><br />David,<br />Thank you for your reply and the link to the paper.<br />Indeed, I fully appreciate your point and I didn't intend to attribute any advantages of the NF thought experiment to you. Indeed I think you brought clarifications to the issue. One thing I'd like to emphasize from my own comment is that, regardless of which expectations formation mechanism you assume, it may actually be the case that, at the ZLB, the NF policy option (central bank announcement on the interest rate + accommodating fiscal policy stance) might result in only modest inflation and a significant output effect: in fact, there would no NF result at all in this case. Besides, if the underlying transmission mechanism is correct, whereby the fiscal impulse leads to higher interest rate and a non-zero output effect, any expectations that would assume a one-for-one increase in nominal interest rate and inflation would not at all be rational, after all. Unless, of course, people believe that the economy is at a new (lower) potential output level but then, again, why would the central bank and the fiscal authority embark on a demand expansionary policy strategy, to start with...?! Thanks againDavid Andolfattohttps://www.blogger.com/profile/12138572028306561024noreply@blogger.comtag:blogger.com,1999:blog-8702840202604739302.post-18153495896908045162015-11-08T14:04:43.105-08:002015-11-08T14:04:43.105-08:00Biago, thanks for your comments.
First, this was...Biago, thanks for your comments. <br /><br />First, this was not a post about any advantages of the Neo-Fisherian thought experiment of raising interest rates. It was just a post to see whether RE was necessary for the result.<br /><br />Having said this, I think I agree with most of your points about the desirability of raising rates in present circumstances. I talk a bit about this near the end of this paper: https://research.stlouisfed.org/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-77915816156836428582015-11-08T13:20:13.335-08:002015-11-08T13:20:13.335-08:00Thanks for this new, clarifying post.
It seems c...Thanks for this new, clarifying post. <br /><br />It seems clear that the first stripe of the Neo-Fisherian proposition is simply untenable: it is a case of 'immaculate' inflation, one where there is no transmission mechanism whatsoever that explains how higher nominal interest rate causes higher inflation (I discuss this issue in my recent comment (http://www.economonitor.com/blog/2015/11/krugman-summers-and-secular-stagnation/)<br /><br />The second stripe, the one that you consider as more plausible, rests on active fiscal policy, without which there would neither be higher nominal interest rates nor higher inflation.<br /><br />But I would like briefly to consider the policy implications of this analysis. First, I would assume that we are in a ZLB world, where conventional monetary policies are ineffective and, thus, the central bank announces a higher nominal interest rate, in anticipation that the fiscal authority accommodates its policy and increases the rate of growth of total public debt: people will rationally expect higher inflation and higher inflation will follow (rightly as you note, not because people expect it, but because the fiscal authority delivers it).<br />Two observations here: <br /><br />First, since we are at the ZLB, resource unemployment is is presumably large: why then should the fiscal impulse translate into higher inflation rather than larger output growth or a combination of both? If the fiscal impulse (partly) triggers output growth, then the nominal interest rate and inflation would not rise one-for-one. The Neo-Fisherian (strip two) result thus holds only at full employment, but why should the central bank want to adopt a Neo-Fisherian policy option and announce a higher nominal interest rate at full employment?!<br /><br />Second, still at ZLB, why should the central bank prefer to resort to the Neo-Fisherian policy option instead of agreeing with the fiscal authority on an overt monetary financing of the deficit (or helicopter money) program, whereby a given fiscal impulse would be fully financed with money creation? This way, the nominal interest rate would be held constant, there would be no new public debt creation, and the output multiplier effect would be larger. <br /><br />In the end, what is the advantage of the Neo-Fisherian policy option? Biagio Bossonehttps://www.blogger.com/profile/12192990143588037522noreply@blogger.comtag:blogger.com,1999:blog-8702840202604739302.post-4576910193213585212015-11-08T06:34:27.606-08:002015-11-08T06:34:27.606-08:00Nick, this looks very, very interesting. Can you p...Nick, this looks very, very interesting. Can you please email me: david.andolfatto@stls.frb.org. Thanks. David Andolfattohttps://www.blogger.com/profile/12138572028306561024noreply@blogger.comtag:blogger.com,1999:blog-8702840202604739302.post-52463175035855463312015-11-08T02:18:16.414-08:002015-11-08T02:18:16.414-08:00David,
I did an exercise earlier this year where ...David,<br /><br />I did an exercise earlier this year where I rewrote the Diamond growth model into an SFC format. (http://monetaryreflections.blogspot.co.uk/2015/02/an-sfc-version-of-diamond-growth-model.html and http://monetaryreflections.blogspot.co.uk/2015/02/national-debt-in-sfc-version-of.html)<br /><br />With suitable choice of parameters, this gives the same results as Diamond, but the different structure allows us to look at some questions slightly differently. The format requires a nominal interest rate, which can then be set as a policy tool. Not surprisingly, with rational expectations (which we need to mirror Diamond), the model is fully neo-Fisherian and the only effect of changing the nominal interest rate is a corresponding change in the inflation rate.<br /><br />However, setting out the model this way makes it easy to replace the RE assumption with adaptive expectations. This still gives the NF result, although it now takes a few periods to get there. (The rate of adaption can't be too quick or the model is not stable.)<br /><br />I think this is quite closely related to what you are saying here. The OLG structure obviously implies a consumption function of the form you describe.<br />Nick Edmondshttps://www.blogger.com/profile/15342983814699700396noreply@blogger.comtag:blogger.com,1999:blog-8702840202604739302.post-80990216277074989822015-11-06T21:33:49.589-08:002015-11-06T21:33:49.589-08:00Yes, this sounds exactly right. Alternatively, it ...Yes, this sounds exactly right. Alternatively, it works also if R = Rm, which is arguably closer to the present situation. The interesting thing here, I think, is that in this scenario, the inflationary consequences of increasing the interest rate follows even in the absence of rational expectations. David Andolfattohttps://www.blogger.com/profile/12138572028306561024noreply@blogger.comtag:blogger.com,1999:blog-8702840202604739302.post-6331657080473128192015-11-06T19:09:12.647-08:002015-11-06T19:09:12.647-08:00David: Good post. I think we are converging.
Here...David: Good post. I think we are converging.<br /><br />Here is perhaps a simpler way to think about it.<br /><br />Start with an almost standard money demand function: M/P = L(Y, R-Rm), L1 > 0, L2 < 0, where Rm is the rate of interest paid on money, and R is the rate of interest paid on other assets, so R-Rm is the opportunity cost of holding money. (The only difference between this and the standard money demand function is that the standard money demand function assumes Rm=0).<br /><br />Now assume that interest on money is paid for by printing new money (so 5% interest on money is like an ongoing annual 1.05 for 1 stock split). This means that printing money to pay interest on money has no fiscal consequences.<br /><br />From then on, it's straight Quantity Theory. Increasing Rm *means* increasing the growth rate of M by exactly the same amount. So it's inflationary.<br /><br />I did a post on it once: http://worthwhile.typepad.com/worthwhile_canadian_initi/2014/08/if-new-money-is-always-paid-as-interest-on-old-money.html<br />(Maybe ignore the last line in that post?)Nick Rowehttps://www.blogger.com/profile/04982579343160429422noreply@blogger.com