tag:blogger.com,1999:blog-2499715909956774229.post1943297469104087410..comments2024-03-22T22:37:02.639-07:00Comments on Stephen Williamson: New Monetarist Economics: Ben Bernanke and the Term Structure of Interest RatesStephen Williamsonhttp://www.blogger.com/profile/01434465858419028592noreply@blogger.comBlogger26125tag:blogger.com,1999:blog-2499715909956774229.post-38216077900096230712013-03-24T08:12:13.124-07:002013-03-24T08:12:13.124-07:00Fiscal policy can do it. This is an old paper, but...Fiscal policy can do it. This is an old paper, but it gives you some insight into this. It's the working paper version:<br /><br />http://www.minneapolisfed.org/research/wp/wp158.pdf<br /><br />By the way, why do you think 10% inflation would solve our problems?Stephen Williamsonhttps://www.blogger.com/profile/01434465858419028592noreply@blogger.comtag:blogger.com,1999:blog-2499715909956774229.post-65391898595249286392013-03-23T23:21:19.509-07:002013-03-23T23:21:19.509-07:00How does Zimbabwe create inflation? Can we copy th...How does Zimbabwe create inflation? Can we copy them, except at 10 percent a year rather than 10 percent a day? Benjamin Colehttps://www.blogger.com/profile/14001038338873263877noreply@blogger.comtag:blogger.com,1999:blog-2499715909956774229.post-37327446171400778132013-03-23T08:05:27.494-07:002013-03-23T08:05:27.494-07:00"...but bank reserves are not federal debt.&q..."...but bank reserves are not federal debt."<br /><br />No. What matters is the consolidated debt of the Fed+Treasury.Stephen Williamsonhttps://www.blogger.com/profile/01434465858419028592noreply@blogger.comtag:blogger.com,1999:blog-2499715909956774229.post-83773847134872250502013-03-23T04:13:56.480-07:002013-03-23T04:13:56.480-07:00Yes, but bank reserves are not federal debt. So wh...Yes, but bank reserves are not federal debt. So when the Fed buys Treasuries with digitized cash, the federal debt is reduced. <br /><br />Also, I have a question. Okay, say I own a $10k T-note. I sell it, and get cash, and the outfit i sell it to sells it to the Fed.<br /><br />How is this not putting cash into the economy? I may put all the money in the bank, but what if I spend some or invest in stocks or property?<br />Benjamin Colehttps://www.blogger.com/profile/14001038338873263877noreply@blogger.comtag:blogger.com,1999:blog-2499715909956774229.post-33466497128988732652013-03-21T21:55:46.895-07:002013-03-21T21:55:46.895-07:00This comment has been removed by the author.Benjamin Colehttps://www.blogger.com/profile/14001038338873263877noreply@blogger.comtag:blogger.com,1999:blog-2499715909956774229.post-77173441695553964972013-03-21T12:41:37.270-07:002013-03-21T12:41:37.270-07:00Not sure if you realize I meant representative age...Not sure if you realize I meant representative agent modelling in the complexity theory sense. The intention of these models is to capture heterogeneity. Anon1noreply@blogger.comtag:blogger.com,1999:blog-2499715909956774229.post-13194052449255008372013-03-21T08:33:32.219-07:002013-03-21T08:33:32.219-07:00I agree with most of what you're saying. Under...I agree with most of what you're saying. Under a floor system (excess reserves outstanding overnight) the IOER determines the overnight rate. But there's some slippage in terms of the effect of the IOER on all other short rates, due to: (i) GSEs; (ii)limitiations on the who can participate in the exchange of reserve account balances. But, (a) I'm not as skeptical as you about our ability to model these things. (b) In some cases these interest rate margins aren't a big deal in terms of how monetary policy works. For example, I think that as IOER rises, that the interest rate differentials - e.g. between IOER and T-bills - should stay roughly constant.Stephen Williamsonhttps://www.blogger.com/profile/01434465858419028592noreply@blogger.comtag:blogger.com,1999:blog-2499715909956774229.post-4337349853894621612013-03-21T06:44:50.659-07:002013-03-21T06:44:50.659-07:00stephen on the front end treasury curve, the reaso...stephen on the front end treasury curve, the reason say 1yr and in is consistently inside 25 bps is because you have a bank that has put a ceiling on that level via it's ballooned reserve balances. (they would always buy above that) while you also have a whole host of people that can't participate in IOER b/c they aren't banks but they have large cash balances, GSE's for one are monster participants in front end cash management they alone can move the front rates 5-10bps just based on their daily coupon inflows/outflows. add in all the other front end cash management out there that can't participate in IOR and you have the reason the front end trades below 25bps. i'm skeptical of any model in all this, it's impossible to capture all the actors involved and their rationale. cidielhttps://www.blogger.com/profile/16087543940772252902noreply@blogger.comtag:blogger.com,1999:blog-2499715909956774229.post-57028460365820758402013-03-21T06:40:51.861-07:002013-03-21T06:40:51.861-07:00Sorry, but the reason I'm not replying to your...Sorry, but the reason I'm not replying to your comments in detail, and just making jokes, is that you are very confused. I would need to teach you a whole course to get you up to speed. But let's start with a small thing:<br /><br />"So the Fed can buy all outstanding US Treasuries, and wipe out the national debt..."<br /><br />No, the Fed cannot "wipe out" debt. A central bank can only change the composition of the outstanding government debt. For example, if the Fed buys T-bills with reserves, what changes, in terms of the outstanding nominal debt? Nothing. Of course, in normal circumstances, we might expect that to increase prices, so the real quantity of outstanding debt would fall. Of course, it's not normal circumstances now - if the Fed swaps reserves for T-bills, absolutely nothing happens, as the T-bills and reserves are identical, to a first approximation.Stephen Williamsonhttps://www.blogger.com/profile/01434465858419028592noreply@blogger.comtag:blogger.com,1999:blog-2499715909956774229.post-44747907651608834832013-03-21T06:33:12.563-07:002013-03-21T06:33:12.563-07:00Benjamin, to be sure, what the fed is doing right ...Benjamin, to be sure, what the fed is doing right now is a tax on savings, part of the theory is to drive investment into riskier parts of the economy by making safe yields less attractive or negative on a real yield basis. right now 10yr real yields are -60bps, given a fairly benign 10yr inflation rate of 2.5%.. as for the fed wiping out the national debt, the answer is no, the fed could wipe out the interest we pay on the debt, it doesn't wipe out the debt unilaterally. if they bot every tsy out there in existence then yes bank reserves just swell and more interest income gets drained from the economy and real yields go even more negative, again tax on savers. banks don't lend reserves they lend TCE ratios etc etc.. the true 'money creators' are the banks from actual loan creation, which again is a function of their overall health not how many reserves they have. hope that clarifies some things, or maybe it makes it muddier!cidielhttps://www.blogger.com/profile/16087543940772252902noreply@blogger.comtag:blogger.com,1999:blog-2499715909956774229.post-78770338440225571772013-03-21T06:32:32.414-07:002013-03-21T06:32:32.414-07:00I agree with you that there are frictions out ther...I agree with you that there are frictions out there that we want to model, and that we want to understand the implications of those frictions for asset prices and for how financial markets function. I think, however, that representative agent models are not the way to go on this. We need to be thinking about financial intermediation and credit market frictions, and heterogeneity is necessary to capture those things.Stephen Williamsonhttps://www.blogger.com/profile/01434465858419028592noreply@blogger.comtag:blogger.com,1999:blog-2499715909956774229.post-75602880956821774932013-03-21T06:29:30.932-07:002013-03-21T06:29:30.932-07:00Benjamin,
"What would happen if they did thi...Benjamin,<br /><br />"What would happen if they did this? You are saying, "nothing. only that bank reserves would swell." Can this be right?"<br /><br />Yes. They'd have to pay IOR at the policy rate in order to raise the rate above zero. Which they'll have to do when we start to get some inflation. Reserves are then the same as t-bills. So they've swapped bonds for t-bills. That's all. Whatever. Khttps://www.blogger.com/profile/09226058602565040485noreply@blogger.comtag:blogger.com,1999:blog-2499715909956774229.post-29683160581490320972013-03-21T05:16:16.097-07:002013-03-21T05:16:16.097-07:00Clarification:
"then removing more of it wil...Clarification:<br /><br />"then removing more of it will increase variance in the market portfolio which will cause it [*the market*] to decline given a constant level of risk aversion."Khttps://www.blogger.com/profile/09226058602565040485noreply@blogger.comtag:blogger.com,1999:blog-2499715909956774229.post-3210766334315393162013-03-21T04:23:20.106-07:002013-03-21T04:23:20.106-07:00Stephen,
Basically I agree with you on all points...Stephen,<br /><br />Basically I agree with you on all points, though I am skeptical that even the signalling channel is very productive. The question is whether the Fed is sufficiently worried about losses to err on excessively stimulative policy for too long. Personally, I think their credibility hinges more on inflation than remittances to the treasury. <br /><br />Here's what bothers me though. Lets assume their portfolio balance theory is correct, and also that their assessment of the risk premium is correct. If the 10-year risk premium is negative that means that the market assesses that it is risk *reducing* to the market portfolio. If that is the case, then removing more of it will increase variance in the market portfolio which will cause it to decline given a constant level of risk aversion. <br /><br />My personal suspicion is that Wallace irrelevance fails, probably due to inconsistent expectations, and central bank purchases can have a significant impact on prices (see e.g. exchange rate targets). I'm pretty sure that if, for example, the Fed announced a $Tn stock purchase we'd get a decent rally. But if buying treasuries is equivalent to *selling* stocks then it's a really bad idea. Maybe that's why stocks fell on both Operation Twist announcements.<br />Khttps://www.blogger.com/profile/09226058602565040485noreply@blogger.comtag:blogger.com,1999:blog-2499715909956774229.post-50677198053353194812013-03-20T22:17:59.200-07:002013-03-20T22:17:59.200-07:00Cidiel, and Stephen Williamson:
I like the idea o...Cidiel, and Stephen Williamson:<br /><br />I like the idea of a Constitutional amendment bearing my name.<br /><br />But let me ask this: Okay, the Fed's QE program does nothing. So the Fed can buy all outstanding US Treasuries, and wipe out the national debt (presuming they hold to maturity)?<br /><br />What would happen if they did this? You are saying, "nothing. only that bank reserves would swell." Can this be right? <br /><br />Sounds like the opportunity of a lifetime to me.Or several lifetimes. <br /><br />BTW, I am asking these questions earnestly. <br /><br />When I say "fresh new money" I mean newly created currency. I am sorry if i do not know the proper terminology. If we increase the amount of currency floating around, that will have no effect? Or a positive effect? Seems like it has to be positive. <br /><br />These seem like reasonable questions to me. <br /><br />Benjamin Colehttps://www.blogger.com/profile/14001038338873263877noreply@blogger.comtag:blogger.com,1999:blog-2499715909956774229.post-6918261243160714272013-03-20T18:09:49.798-07:002013-03-20T18:09:49.798-07:00To say what the data means, I have to have an expl...To say what the data means, I have to have an explicit theory. So, in this instance I would like to have a theory that prices bonds of different maturities, and the theory should allow me to say how different factors matter - inflation, monetary policy, for example. It should also tell me about the link between real and nominal rates at different maturities. The problem here is that the theory is not formalized. There is something that is being estimated here, but we have no idea what it is - I need the theory to tell me how to interpret those estimates.<br /><br />But, one thing I know is that, on average, the nominal yield curve is upward-sloping, so it appears that long-maturity nominal government debt sells at a lower price - adjusting for duration - than does short maturity government debt. Thus there is a regularity in the data - short maturity government debt commands a premium in financial markets. Why? No one has successfully explained that. It could be a liquidity premium of some sort, or you could call it a term premium if you want. But without a theory we can't make any sense of it, or measure it.Stephen Williamsonhttps://www.blogger.com/profile/01434465858419028592noreply@blogger.comtag:blogger.com,1999:blog-2499715909956774229.post-2230784484478494162013-03-20T18:02:50.359-07:002013-03-20T18:02:50.359-07:00Agency issues abound in the bond market. These ca...Agency issues abound in the bond market. These can result in homogeneity of duration bet strategies that in turn lead to price distortions. For instance, since '09 the term real yield has declined into negative territory while stock prices have more than doubled. This divergence is difficult to explain under conditions that allow arbitrage. The question is not whether QE "works" as its proponents advertise. Instead, its whether such an outcome can emerge out of a few simple market inefficiencies. A representative agent model would likely capture this dynamic.Anon1noreply@blogger.comtag:blogger.com,1999:blog-2499715909956774229.post-3701019294304428242013-03-20T16:04:38.192-07:002013-03-20T16:04:38.192-07:00I can't tell if you're skeptical about the...I can't tell if you're skeptical about the idea of a term premium or skeptical about the measurement of the term premium.JP Koninghttps://www.blogger.com/profile/02559687323828006535noreply@blogger.comtag:blogger.com,1999:blog-2499715909956774229.post-12129005693086423792013-03-20T15:51:01.448-07:002013-03-20T15:51:01.448-07:00Sure. The Fed said something, and asset prices cha...Sure. The Fed said something, and asset prices changed. So, it's Fed driven, but that's consistent with QE having no direct effect, and just acting as a signal of a change in future Fed actions.Stephen Williamsonhttps://www.blogger.com/profile/01434465858419028592noreply@blogger.comtag:blogger.com,1999:blog-2499715909956774229.post-17910935062936125852013-03-20T12:08:57.923-07:002013-03-20T12:08:57.923-07:00for example 3y5y rates stood in a range between ro...for example 3y5y rates stood in a range between roughly 3.5 to 4.5% for several years including post financial crisis. but suddenly in august 2011 which is when the 'mid 2013' surprise statement came out, that rate has yet to peak above 3% since that time. it rallied 100+ bps in the 10 days after that statement and settled in a range much lower than 3%. I would call that a direct result of an unexpected flattening of future rate expectations, that was primarily fed driven. cidielhttps://www.blogger.com/profile/16087543940772252902noreply@blogger.comtag:blogger.com,1999:blog-2499715909956774229.post-14095217408812651682013-03-20T11:29:25.710-07:002013-03-20T11:29:25.710-07:00"...that the significant drop at the end of 2..."...that the significant drop at the end of 2011 was absolutely fed driven..."<br /><br />How do you know?Stephen Williamsonhttps://www.blogger.com/profile/01434465858419028592noreply@blogger.comtag:blogger.com,1999:blog-2499715909956774229.post-1059478337568297942013-03-20T10:09:06.327-07:002013-03-20T10:09:06.327-07:00Benjamin, I'll start with
"I have a que...Benjamin, I'll start with<br /><br /> "I have a question. Yes, the Fed is an intermediary. But does not the Fed also create money? And when a central bank creates money, does not that count for something?"<br /><br />NO the fed doesn't create money, they create reserves which they swap for interest bearing treasuries, they effectively take interest income out of the economy to hopefully lower borrowing costs and spur the fiscal channel to work.. that's how it's supposed to work at least but for the moment the fiscal channel is officially broken.. i digress, Stephen, another excellent post thx.. and while i agree with many points about the effectiveness of monetary policy and for me this is especially true as time progresses with the same policy. I will say in the chart of deconstructed 10yr yields, that the significant drop at the end of 2011 was absolutely fed driven and can be shown in forward yield curves across the term structure that all dropped roughly 100+bps. This was largely driven by the date driven commitment from the fed to the end of 2014? i believe? It was in that one instance;where the commitment appeared a very effective tool for driving down term interest rates.cidielhttps://www.blogger.com/profile/16087543940772252902noreply@blogger.comtag:blogger.com,1999:blog-2499715909956774229.post-49028375481545288642013-03-20T07:01:39.965-07:002013-03-20T07:01:39.965-07:00"Benjamin was right and deserved a better res..."Benjamin was right and deserved a better response."<br /><br />This is an amendment to the Constitution maybe?Stephen Williamsonhttps://www.blogger.com/profile/01434465858419028592noreply@blogger.comtag:blogger.com,1999:blog-2499715909956774229.post-26225868525465871802013-03-20T05:18:44.379-07:002013-03-20T05:18:44.379-07:00It is asserted, "They need a model," whi...It is asserted, "They need a model," which assumes that a model is possible.<br /><br />What if, based on our lack of knowledge, psychology, bias, etc., any set of facts has 10 or 20 outcomes or 100s of billions?<br /><br />It seems to me that all you have shown is that economics is like quantum mechanics---sometimes it works---but we don't know whether the cat is alive or dead and well never know.<br /><br />Benjamin was right and deserved a better response.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-2499715909956774229.post-21997726854020791162013-03-20T05:09:51.993-07:002013-03-20T05:09:51.993-07:00"This is fresh new money."
Yes, fresh m..."This is fresh new money."<br /><br />Yes, fresh money is the best.Stephen Williamsonhttps://www.blogger.com/profile/01434465858419028592noreply@blogger.com