tag:blogger.com,1999:blog-8702840202604739302.post3219809676474163291..comments2024-03-28T03:38:53.734-07:00Comments on MacroMania: A reply to Lawrence WhiteDavid Andolfattohttp://www.blogger.com/profile/12138572028306561024noreply@blogger.comBlogger14125tag:blogger.com,1999:blog-8702840202604739302.post-30124321224387675192017-03-10T17:12:35.956-08:002017-03-10T17:12:35.956-08:00I omitted my summary in the note above. My princip...I omitted my summary in the note above. My principal point is that interest on reserves policy has shifted the Fed’s focus of monetary policy to credit policy, and the effect of expanding the Fed’s credit has been to repress the banking sector and cause taxpayer losses. Your focus on Fed shift to acquiring risky long-term assets and thereby boosting its revenue is not a meaningful issue. The higher Fed revenue just offsets the higher interest payments with a higher duration premium on longer-term debt. The same is true for MBS that carry a default premium, but the analysis is more complicated to the extent that the issuers are or are not part of the Federal Government. The rise in revenue, like the rise in assets, has been at the expense of the banking system and the efficiency of our financial system.<br /><br />When the Fed conducts monetary policy, the constraint on the size of the Fed’s balance sheet is its inflation objective. The constraint on its balance sheet with QE credit policies is how much financial repression it wants to cause, or the limits on the ability to suck all the risky assets out of depository institutions and on to the Fed's balance sheet. <br />Jack Tatomhttps://www.blogger.com/profile/10985950297639983329noreply@blogger.comtag:blogger.com,1999:blog-8702840202604739302.post-70183072322920030712017-03-10T11:06:58.950-08:002017-03-10T11:06:58.950-08:00I want to question the basic premise, restated in ...I want to question the basic premise, restated in your response to Larry White:<br /><br />“The gist of my story went something like this. Suppose that the Fed makes a one-time purchase of a 10-year treasury bond yielding a risk-free 2.5% annual nominal coupon. Suppose that the purchase is financed by "printing" reserves. (Note, this asset purchase and method of finance describes the basic nature of QE). Suppose further that the interest paid on reserves remains below 2.5% for the duration of the bond. Then the Fed makes a profit off the rate of return differential in each of the 10 years it holds the bond on its books. If I understand my critics correctly, none would dispute the financial benefit associated with this Fed intervention as far as the public purse is concerned; at least, as long as it is somehow known ex ante that the short rate will remain below the long rate in the manner assumed.”<br /><br />First, let the interest rate on reserves be zero. If the Fed bought a T-bill from a bank for $100 that paid 6 basis points, the average on 4-week T-bills from fall 2008 through the end of 2015, or if it bought a $100 10-year Treasury yielding 2.5 percent, the Federal debt outside the Fed would fall by $100 and the monetary base would rise by $100. The Fed would have higher earnings by the amount of the interest, higher in the case of the higher yielding security, but in either case the additional profit would equal the interest paid by the Fed and equal the higher payment from the Fed to the Treasury. A differential effect due to the duration premium would have no effect on the Treasury debt reduction, or on the initial or subsequent changes in the sum of Fed and bank credit, or on money. <br />This is “pre-2008 monetary economics.” <br /><br />Now suppose the Fed pays an interest rate on reserves, say 25 basis points, that exceeds the market rate on comparable term T-bills, 6 basis points. This additional action means that when the Fed conducts open market operations buying T-bills or longer-term Treasury securities that have a comparable risk-adjusted yield, the depository institution will profit by selling the Treasury security and holding excess reserves. No increase in the effective monetary base, currency or required reserves, takes place. The Fed expands its credit without affecting monetary aggregates. In effect the Fed can bail out banks, extending its credit to banks by acquiring questionable assets or more risky assets, without affecting money. Now the money supply does not expand, nor does total credit (Fed plus bank credit). Banks swap credit for excess reserves and the Fed has the T-bills as part of Reserve Bank Credit. Federal debt outside banks still goes down by 100, but the Fed excess reserve offsets this. The Fed owes more annual interest than the Treasury owed on the T-bills. The true fiscal position (on a consolidated Treasury plus Fed basis) worsens rather than improving as in the Open Market Operation described above, and by the extent that the interest rate on reserves exceeds the interest rate on T-bills. <br /><br />This is a bad deal for the taxpayer. The taxpayer now pays 25 basis points for the $100 of excess reserves banks hold instead of the 6 basis points Treasury was paying banks, or its risk adjusted equivalent for a 10-year security. Of course there is no new effective monetary base and no new money or total credit in this case. Fed credit goes up, but bank credit declines by an equal amount. This is called financial repression, which is known to reduce private risk-taking and economic growth. The extra cost to the Fed for excessive interest will soon reach about $20 billion per year when the interest rate on reserves is put up to one percent. <br /> <br />Eliminating interest on excess reserves would solve the balance sheet size problem with little adverse effect so long as Fed assets are liquidated at the same pace. <br />Jack Tatomhttps://www.blogger.com/profile/10985950297639983329noreply@blogger.comtag:blogger.com,1999:blog-8702840202604739302.post-45239833738732883462017-03-06T07:08:04.342-08:002017-03-06T07:08:04.342-08:00In agreement with your risk piece. Why the Fed is...In agreement with your risk piece. Why the Fed is (seemingly) unconcerned with the scenarios you present remains, however, a mystery. Diegonoreply@blogger.comtag:blogger.com,1999:blog-8702840202604739302.post-41554496353064667942017-03-04T11:17:52.053-08:002017-03-04T11:17:52.053-08:00"The Fed has no special "capacity" ..."The Fed has no special "capacity" to remit revenues by taking duration risk as Treasury's fiduciary."<br /><br />As I said in my post, years and years (decades even) of elevated remittances, one fluke after another. No amount of evidence will convince you of my "post hoc fallacy." <br /><br />As for the scenario you'd like me to write about, I don't have the time to do it right now. But I have written about balance sheet risk in the past: http://andolfatto.blogspot.com/2013/03/fed-balance-sheet-risks.htmlDavid Andolfattohttps://www.blogger.com/profile/12138572028306561024noreply@blogger.comtag:blogger.com,1999:blog-8702840202604739302.post-30665314857732197462017-03-04T09:36:10.621-08:002017-03-04T09:36:10.621-08:00David,
"remove its capacity to remit close t...David,<br /><br />"remove its capacity to remit close to $100B..."<br /><br />I think I hear a post hoc fallacy. Observed positive returns are not evidence of the absence of risk (so it says on every mutual fund advertisement). The Fed has no special "capacity" to remit revenues by taking duration risk as Treasury's fiduciary. <br /><br />In any case, why not do a post on a "behind the curve" scenario for the Fed? It raises rates, but inflation rises by more (as you argue, the Fed's models are not predictive...). The Fed is reluctant to catch up as it mistakenly sees inflation as "transient". What would happen to deficit projections as rates and inflation rose together? How would this feed back into inflation expectations? Let's say the Fed caught up, and sparked a downturn. To fight it, it does another $3tr in QE. What would happen to the deficit and inflation in the next cycle with $6tr in T-bill equivalents facing rate rises? I've never heard a Fed speaker or related economist, or a Treasury official, explore the risk of this scenario. It would be refreshing to see it rigorously explored and even dismissed. Since you've stepped into the public finance sphere, you're the logical person to take that on :). Diegonoreply@blogger.comtag:blogger.com,1999:blog-8702840202604739302.post-23121257988514801522017-03-04T08:18:25.234-08:002017-03-04T08:18:25.234-08:00Let's go back to what appears to be your main ...Let's go back to what appears to be your main concern:<br /><br />"If the Fed bets wrong and short rates spike, the deficit will widen from CBO baseline."<br /><br />So, your solution for this "problem" is for the Fed to immediately remove its capacity to remit close to $100B/year to the Treasury--an act that would have an immediate and long lasting impact on the deficit. But this would be OK because now the Treasury doesn't have to worry about Fed remittances moving unexpectedly up and down? I'm afraid I just don't follow this line of reasoning. David Andolfattohttps://www.blogger.com/profile/12138572028306561024noreply@blogger.comtag:blogger.com,1999:blog-8702840202604739302.post-71840145941468280562017-03-04T08:04:03.548-08:002017-03-04T08:04:03.548-08:00David,
..."I'm not sure why they cannot ...David,<br /><br />..."I'm not sure why they cannot do the same with Fed liabilities."<br /><br />Here's a reason:<br />Imagine the Fed sent Treasury an memo: "ICYMI, we've just transformed $3tr of your term debt into T-bill equivalents." For Treasury to offset the duration impact of this, they would have to short $3tr nominal of bond futures. The alternative, exchanging term debt for IOR-paying reserves, is not possible. If they shorted the futures, the Fed could simply send Treasury a second email: "ICYMI, to maintain the desired effect of QE, we went into the bond futures market and took the other side of your recent trade; therefore, your consolidated duration risk remains the same". <br /><br />Any attempt by Treasury to manage the term structure of consolidated (i.e. including Fed balance sheet) liabilities would require the Fed's cooperation; that in turn requires the Fed give up independently conducting QE. Diegonoreply@blogger.comtag:blogger.com,1999:blog-8702840202604739302.post-6174061179028244082017-03-04T07:06:48.830-08:002017-03-04T07:06:48.830-08:00Diego, on your second point, I find it rather wak....Diego, on your second point, I find it rather wak. The Treasury has to manage duration risk of its own debt, I'm not sure why they cannot do the same with Fed liabilities. And how about not counting on high levels of Fed revenue to begin with? Then the Treasury can only be pleasantly surprised. On average, as the data plainly shows, the Treasury receives revenue from the Fed. Why are you so keen for the Treasury to give that up? David Andolfattohttps://www.blogger.com/profile/12138572028306561024noreply@blogger.comtag:blogger.com,1999:blog-8702840202604739302.post-49818746185331961672017-03-03T15:08:20.939-08:002017-03-03T15:08:20.939-08:00I'll add two points to White's arguments. ...I'll add two points to White's arguments. <br /><br />Until 2007 Fed remittances came almost entirely from seigniorage, which is a free lunch. Post 2007, Fed remittances resulted from normal risk-taking. The two are different not in degree, but in kind. <br /><br />Second, the Fed has serious flaws when it comes to government liability management. First, it does not concern itself with the budget. If the Fed bets wrong and short rates spike, the deficit will widen from CBO baseline. There are people at Treasury and at the Treasury Borrowing Advisory Committee (TBAC) that are to manage this type of rate risk. Are you aware of anyone at the Fed that has this specific role, and how many people they manage? The topic seems to be absent from the staff presentation in the FOMC minutes. I would check the org chart! :)<br /><br />Diegonoreply@blogger.comtag:blogger.com,1999:blog-8702840202604739302.post-40095620878356933392017-03-02T12:05:57.173-08:002017-03-02T12:05:57.173-08:00I think this is an example of differences in langu...I think this is an example of differences in language and intuitions. A macro theorist might talk about productivity shocks and their effects on financial market equilibrium. But as a market practitioner, JKH will talk about the tools now in use to manage such risk, which are bonds and interest rate derivatives.<br /><br />The reaction to Robert Shiller included people asking if his innovations are really needed, and what would actually change in the market. I think it would clean up and simplify the macroeconomic risk management system, while also making it more effective, and consuming fewer human resources. We also need to relax important constraints like the ZLB that are connected with the interest-rate management methodology.Anwerhttps://www.blogger.com/profile/08277173974258559733noreply@blogger.comtag:blogger.com,1999:blog-8702840202604739302.post-43007776625742041732017-03-02T10:35:19.248-08:002017-03-02T10:35:19.248-08:00https://www.brookings.edu/blog/up-front/2014/10/01...https://www.brookings.edu/blog/up-front/2014/10/01/should-the-treasury-and-the-fed-cooperate-on-debt-management/<br /><br />That's the link to the debt management panel at Brookings.Anonymoushttps://www.blogger.com/profile/14248645887343991539noreply@blogger.comtag:blogger.com,1999:blog-8702840202604739302.post-6248929920205633252017-03-02T10:33:45.573-08:002017-03-02T10:33:45.573-08:00Dr. A, you might enjoy the Brookings panel in whic...Dr. A, you might enjoy the Brookings panel in which Larry Summers points out that the Treasury lengthening term and the Fed shortening term have been at cross purposes. Summers et al argue that we might take out the middle man, Fed in this case, and just issue all debt as bills. That really got the Treasury official on the panel squirming. Sort of the anti-Cochrane. JC likes a big Fed balance sheet for banking system soundness reasons but wants the debt in perpetuals, I think.<br /><br />I notice your argument has little to do with "stimulus" or with trying to manipulate the term structure. As I read you, the Fed -- acting for Treasury -- might just respond to the term structure that exists. It is quite different from the standard argument for QE. I do not mean that as criticism. Just joining the conversation. Thanks. (I will come back with the link.) Anonymoushttps://www.blogger.com/profile/14248645887343991539noreply@blogger.comtag:blogger.com,1999:blog-8702840202604739302.post-80061825083836437382017-03-02T03:44:41.545-08:002017-03-02T03:44:41.545-08:00The interesting question concerns how Treasury doe...The interesting question concerns how Treasury does in fact manage the maturity structure of its debt. I’ve not seen much written on this (although I haven’t looked very hard).<br /><br />For one thing, given that fiscal management and planning would rationally consider a forecast for Fed profit remittances, which in turn must consider Fed balance sheet structure, it should be the case that Treasury does in some way manage the consolidated maturity structure – in the sense that it should be responding in real time (at least fiscal year time) to the expected marginal risk/return effect of QE type interest rate transformations undertaken by the central bank, along with everything else it needs to consider. So the “already optimal” point is a bit tricky I think.<br /><br />The question concerns risk management. Commercial banks got slammed by interest rate risk in the 1980’s. One obviously has to consider the difference when comparing that with the general interest rate environment today, as well as the difference between government financial risk management and private sector financial risk management (e.g. the usual solvency economics comparison issue, the fact that the central bank doesn't need to take unrealized marked to market losses on Treasuries into a recognized capital position, etc. etc.) So there are important mitigating factors from that starting point.<br /><br />But risk is risk. And in both cases one needs to consider the difference between economic present value risk and interest margin risk. Notwithstanding the very good bet that “short funding” interest rate risk generally works well through the long run (as indicated by your graph - with the 1980’s being the most concerning in that regard), the potential for stomach churning interest rate resets in a rising rate environment still needs to be considered – for example, in the extreme case where the entire consolidated federal debt is short funded. This isn’t just political. It’s a matter of recognizing risk and how it could affect the budget in an unexpected way, notwithstanding the reasonable bet that this can easily be absorbed over the long run. And if it’s imprudent to fund the entire federal debt on that basis because of that, then limits need to be determined.JKHhttps://www.blogger.com/profile/06322177539880818556noreply@blogger.comtag:blogger.com,1999:blog-8702840202604739302.post-52318564367031574352017-03-01T14:26:23.073-08:002017-03-01T14:26:23.073-08:00Some might have very good intuition for the concep...Some might have very good intuition for the concepts and language Larry uses ("duration risk") but I like to start with the question of who has a comparative advantage in bearing the risk of productivity shocks. This is basically the question that Robert Shiller engaged in his advocacy of NGDP-linked sovereign debt, suggesting that countries should shift this risk to the rest of the world.<br /><br />The US doesn't issue such debt, but it does participate in a risk-management regime, involving banks and others that hold bonds. When necessary, the Fed lowers rates, and this recapitalizes banks and provides capital appreciation to other users of such bonds. Markus Brunnermeier is especially well-known for discussion of this redistributive type of monetary policy. One may ask if it's time to move to a cleaner risk-management implementation that optimizes for multiple objectives, including the liquidity satiation advocated by Milton Friedman (which you mention in your post). For example, if the US issues all debt in perpetual form, with GDP deflator as the accounting unit, then this is money and it also stabilizes the US budget in a way that allows for tax smoothing.<br /><br />But that would be a big change in the system, perhaps too radical for US officials to contemplate seriously.Anwerhttps://www.blogger.com/profile/08277173974258559733noreply@blogger.com