tag:blogger.com,1999:blog-6837159629100463303.post1130244595541134280..comments2023-06-18T01:25:08.748-07:00Comments on Information Transfer Economics: Who should we listen to?Jason Smithhttp://www.blogger.com/profile/12680061127040420047noreply@blogger.comBlogger5125tag:blogger.com,1999:blog-6837159629100463303.post-6183558760133646732015-11-04T09:06:03.529-08:002015-11-04T09:06:03.529-08:00Hi LK,
I have no problem with the claim that &quo...Hi LK,<br /><br />I have no problem with the claim that "dollar for dollar, monetary injections have less impact", but at that point it becomes semantics about the liquidity trap, monetary offset and fiscal policy. Saying monetary policy has less impact and saying an economy is in a liquidity trap are just a difference in degree, not kind.<br /><br />You said<br /><br /><i>"However, I think Scott Sumner rightly claims that the velocity does not fall exactly in sync with increases in the base."</i><br /><br />First, I think Sumner considers the equation of exchange (an particularly velocity) to be a definition:<br /><br /><a href="http://www.themoneyillusion.com/?p=28712" rel="nofollow">http://www.themoneyillusion.com/?p=28712</a><br /><br />"[MV = PY] means V is PY/M. And that’s all it means."<br /><br />So it's not really an explanation, but rather a name for the mis-match between PY and M. He does go on in that post to describe an interpretation of 1/V = k as the fraction of NGDP held as cash balances. But again, that is just an accounting identity (definition).<br /><br />You asked:<br /><br /><i>"That said, what is the IT model prediction for the relationship between injections to the base and NGDP?"</i><br /><br />The monetary base has a direction relationship on short term interest rates in the IT model; it's described in the paper:<br /><br /><a href="http://informationtransfereconomics.blogspot.com/2015/08/information-equilibrium-as-economic.html" rel="nofollow">http://informationtransfereconomics.blogspot.com/2015/08/information-equilibrium-as-economic.html</a><br /><br />Increases in physical currency (monetary base minus reserves) appears to have a direct relationship with NGDP<br /><br /><a href="http://informationtransfereconomics.blogspot.com/2014/03/the-monetary-base-as-sand-pile.html" rel="nofollow">http://informationtransfereconomics.blogspot.com/2014/03/the-monetary-base-as-sand-pile.html</a><br /><br />where the difference appears to derive from labor market fluctuations:<br /><br /><a href="http://informationtransfereconomics.blogspot.com/2015/08/employment-doesnt-depend-of-inflation.html" rel="nofollow">http://informationtransfereconomics.blogspot.com/2015/08/employment-doesnt-depend-of-inflation.html</a><br /><br /><i>"Also, I understand that it allows for phase transitions out of its take on 'secular stagnation'. Is that right? How does this work, exactly?"</i><br /><br />Secular stagnation is not directly related to the "phase transitions" (although a phase transition may get you out of secular stagnation). I referred to the changes as both phase transitions and monetary regime changes. They would find a reasonable description in terms of expectations. Nick Rowe has used the analogy of driving on the right. In the US and Canada, we drive on the right because of expectations anchored by the government. A 'phase transition' would occur if expectations changed and everyone started driving on the left. That is one way to look at monetary regime change in the IT model -- a certain set of expectations hold and then the central bank changes the way it conducts monetary policy and expectations change.<br /><br />This is not a well-understood aspect of the IT model (I don't know what constitutes a monetary regime change). 'Phase transition' is just a name for an apparent effect. I think of them as monetary regime changes because they seem to coincide with major changes in monetary policy in the UK and the US (and a less significant change in Switzerland).<br /><br />Regarding secular stagnation itself, it is a result of entropy -- ignorance about the micro states of the economy. If I don't know how each dollar of money is allocated against each dollar of NGDP, then I must assume ignorance and therefore a maximum entropy allocation (distribution). If you do this, there are more ways a large economy can be organized as a collection of many low growth states than a small economy (which has a higher probability to be organized as a few high growth states). Therefore, large advanced economies tend to have slower growth on average than small developing ones.Jason Smithhttps://www.blogger.com/profile/12680061127040420047noreply@blogger.comtag:blogger.com,1999:blog-6837159629100463303.post-36842469826074521422015-11-04T06:42:14.713-08:002015-11-04T06:42:14.713-08:00"Its effectiveness suddenly goes from about 1..."Its effectiveness suddenly goes from about 1 - 1.3 to 1 to 10 to 1 in 2009. That seems very much like the liquidity trap argument. Scott Sumner wouldn't like that at all."<br /><br />I won't speak for Scott Sumner. My take is that there is certainly a form of liquidity trap going on; dollar for dollar, monetary injections have less impact (i.e. velocity slows down rapidly at each injection). However, I think Scott Sumner rightly claims that the velocity does not fall exactly in sync with increases in the base. Even in a liquidity trap, injections have an effect. Ergo, monetary policy can be effective.<br /><br />As for the relationship between growth and the base, Sadowski looked at impulses from a baseline. If I understand, the model essentially starts from the hypothesis there would be some baseline NGDP growth without injections. Then, adding or subtracting to the base (in the current macro environment) has an effect of 10:1.<br /><br />By the way, I think this kind of empirical analysis is useful to establish a rule of thumb, but lacks, for example, an understanding of the mechanics that brought us into the liquidity trap or how/when it can be escaped. You entropy-based model, in my opinion, is much better, from that point of view.<br /><br />That said, what is the IT model prediction for the relationship between injections to the base and NGDP? Also, I understand that it allows for phase transitions out of its take on "secular stagnation". Is that right? How does this work, exactly? I am quite curious.LK Bélandnoreply@blogger.comtag:blogger.com,1999:blog-6837159629100463303.post-4834182104580580602015-11-03T21:07:09.623-08:002015-11-03T21:07:09.623-08:00Scott Sumner's model is that QE is expected to...Scott Sumner's model is that QE is expected to be taken away, therefore it does not generate inflation, so actually, that raises more questions.<br /><br />Between 1980 and 2000, the monetary base grows about 470%. That would imply by the 10:1 measure that NGDP should grow 47%. But NGDP grows 360% during that time. A 1.3:1 ration.<br /><br />Between 1980 and 1990 the monetary base grows about 210%, The 10:1 measure would mean NGDP grows 21%, but it grows 210% (a 1:1 ratio).<br /><br />That means monetary policy seems to be getting less effective over time. Scott Sumner doesn't think that is true. <br /><br />Its effectiveness suddenly goes from about 1 - 1.3 to 1 to 10 to 1 in 2009. That seems very much like the liquidity trap argument. Scott Sumner wouldn't like that at all.<br /><br />So is it just excess reserves that leads to more output (therefore monetary policy works differently before vs after 2009)? Scott Sumner wouldn't agree with that.<br /><br />Overall Sadowski's argument is that we have to double the base (100% increase) to get 10% NGDP over 5 years. (I think it was inflation in the original posts.) Therefore we'd have to double it every 5 years to maintain 2% NGDP growth. So over the next 50 years, the base would grow 1024 = 2^10 times, while NGDP would grow 20% = (1.02)^10. The base would be 4 quadrillion dollars while NGDP would be 22 billion dollars.Jason Smithhttps://www.blogger.com/profile/12680061127040420047noreply@blogger.comtag:blogger.com,1999:blog-6837159629100463303.post-6411580294462346392015-11-03T20:57:58.026-08:002015-11-03T20:57:58.026-08:00Excellent blogging. You could square Sumner's...Excellent blogging. You could square Sumner's circle by pointing out Sumner is setting up a series of strawmen (as is his wont) that represent wrong views, with only his view (#10) being the correct one. Sumner is entertainment, but in a world where people don't want to do hard science (and that includes me, I've made much more money in business than in science, and I have several degrees in science) Sumner is a convenient 'brand name' to identify NGDPLT with, hence his blog gets read more than this one, though this one is more scientific by far.Ray Lopezhttps://www.blogger.com/profile/11134761834999705305noreply@blogger.comtag:blogger.com,1999:blog-6837159629100463303.post-38890573746122969112015-11-03T18:24:29.984-08:002015-11-03T18:24:29.984-08:00Didn't Mark Sadowski's series of VAR posts...Didn't Mark Sadowski's series of VAR posts address many of your issues with Scott Sumner's claims? Basically, a 10% expansion in the Fed's balance sheet leads to 1% more NGDP growth.<br /><br />He also looked at similar effects on exchange rates, the stock market and other variables of interest.LK Bélandnoreply@blogger.com