Tuesday, June 29, 2010

Fed Term Deposit Facility

The Fed conducted its first auction of funds through its new term deposit facility on Monday, as described here. Recall that the term deposit facility allows banks with reserve accounts to effectively tie up reserve funds in an account that cannot be used in intraday transactions and cannot count towards overnight reserve requirements. The idea, as I have discussed before (though I couldn't find the relevant blog piece) is that this promotes inflation control by somehow tying up reserves so that they cannot "escape." I have also argued that this is misguided - the Fed can achieve all the inflation control it needs, and at lower cost, by setting the interest rate on reserves appropriately.

Now, what happened in the auction? $2 billion in 28-day term deposits were offered, and an amount greater than $11 billion was the amount tendered. I have no idea how an auction like this might go, but this looks like healthy demand, though of course you would have to see all the prices that were associated with the $11 billion quantity to determine that. The "stop out rate" was 0.27%, which is the lowest accepted bid rate. Since the interest rate on reserves is currently 0.25%, the Fed does not have to pay much of a premium to get a bank to commit to locking up reserves for 28 days. Of course this is due to the fact that there is a huge quantity of reserves in the system, so a marginal unit of reserves has close to zero value in making intraday transactions or in satisfying reserve requirements. Thus, these term deposits are not costing the Fed much more than liquid reserves, though presumably it costs something to run the auction. One would expect though, that as the quantity of reserves falls, the margin between the term deposit rate and the interest rate on reserves will rise.

Correction: This was not the first auction. That was actually in mid-June. See this.

14 comments:

  1. The media has been saying these term deposits will act as a floor for the interest rate. You don't think this will happen though?

    Also I'm wondering about the extraordinary circumstances we're in with the very weak real estate market. Using these deposits the Fed can keep interest rates low while still containing inflation with term deposits. The low interest rates would be beneficial in sustaining home prices (assuming that when consumers spend less on interest they simply spend more in the principal). Your thoughts?

    ReplyDelete
  2. You wrote:
    "One would expect though, that as the quantity of reserves falls, the margin between the term deposit rate and the interest rate on reserves will rise."

    This is wrong. When private banks want to park their excess liquidity risk free for the short term, they have three fundamentally equivalent options: one-month T-Bill, excess reserves or term deposit facility. Arbitrage possibility between these three options will make sure that the interest rate paid on excess reserves de facto set the interest on 1 month T-Bill and 1-month term deposit facility (as long as traders don't expect the Fed fund rate to go up in the next month, which is obviously the case right now).

    In a world with no excess reserves (like prior to the crisis), there is no point in having term deposit facility. So the only remaining option for private banks to park their liquidity risk-free in the very short term is T-Bills.

    Qc

    ReplyDelete
  3. Wayfarer: No, under the current circumstances (positive quantity of excess reserves in the system) the interest rate on reserves is the relevant "floor." Note of course that both T-bill rate and the effective fed funds rate are lower than the interest rate on reserves. There are specific reasons for that, but roughly it is the interest rate on reserves that determines all short-term rates right now. Obviously the interest rate on reserves can't go much lower than it is (0.25%).

    Anonymous: I think we have had this discussion before. T-bills, excess reserves, and term deposits are not quite equivalent. The difference has to do with liquidity. T-bills are more liquid than reserves currently, as some financial institutions cannot hold reserve accounts, and T-bills are widely used as collateral. Term deposits are less liquid than reserves as they cannot be used in daylight transactions, but as I argued in the piece that doesn't matter much at the margin right now. As reserves become more scarce, it will.

    ReplyDelete
  4. Ok, when is the last time you saw the one-month T-Bill depart significantly from the Fed Fund rate? You might want to draw a big distinction from a liquidity perspective between these three risk-free investment vehicle, but the proof is in the pudding... yields on these investments are and will always be very similar.

    Also, I am happy to see that you are now in full agreement with me that excess reserves are no more inflationary than T-Bills, and that you have dropped the idea that banks might want at some point to get rid of their excess reserves which could lead to inflation:
    "The idea, as I have discussed before (though I couldn't find the relevant blog piece) is that this promotes inflation control by somehow tying up reserves so that they cannot "escape." I have also argued that this is misguided - the Fed can achieve all the inflation control it needs, and at lower cost, by setting the interest rate on reserves appropriately."

    I realise -knowing well the natural and gigantic egos prevailing in academic circle- that you can not acknowledge that an anonymous blogger has changed your views on excess reserves and their impact on inflation.

    In any case, this was my last post on your blog, I did enjoy exchanging with you.

    Qc
    BTW- I am absolutely not an "old monetarist"... On other blogs I have also been called a "post-keynesian" or a "Lerner disciple" although I have never read a single monetarist, post-keysesian or Lerner paper in my entire life. I am just fascinated with money creation, banking and fiscal operations. That is all.

    ReplyDelete
  5. I'm glad to hear we agree. I hope you don't think of me as a gigantic-ego type. By nature I'm rather modest, but I've acquired some aggressive characteristics, mainly in order to survive in this nasty profession.

    ReplyDelete
  6. Stephen Williamson,

    Sorry to thread hijack, but what is your opinion on the State of Macroeconomics post by Morley from your university. He is quite critical of the DSGE approach.

    ReplyDelete
  7. Yes, I read it. James is in fact leaving to go to UNSW in Sydney. James does not talk to the macroeconomists in the department, and I had no knowledge that he was writing this piece. It's basically a forecaster's view of the state of the art in macroeconomics. James is an excellent forecaster, but he is a little confused about who is doing what in macro and why. James thinks of "DSGE" as including real business cycle (RBC) models and modern estimated New Keynesian models (Smets-Wouters for example). New Keynesians seem to have appropriated "DSGE" to describe what they do, but basically this class of models is anything from the 3-equation form (IS, Phillips curve, Taylor rule) to large models of the type that Larry Christiano works with. James differentiates what he does (large-scale macroeconometric forecasting models) from what he calls DSGE (RBC and actual DSGE), but for me these models are essentially all the same breed. From my point of view, New Keynesian DSGE models have evolved to the point where they are very much like the old large-scale macroeconometric models used in the 1970s, for example. The computation that goes into estimating and solving the modern models is much more sophisticated than it was in 1970, and these models fit the data well, but I wouldn't trust the policy implications that come out of them.

    ReplyDelete
  8. "the Fed can achieve all the inflation control it needs, and at lower cost, by setting the interest rate on reserves appropriately"

    This "lower cost" argument can be also applied to the activities of treasury. Why are they bothering with 30 year treasuries? 3 month bills would be cheaper.

    ReplyDelete
  9. Excellent. That has always puzzled me too. Maybe someone has a theory about this. Does anyone know? Even now, the government can borrow for 30 years at at little less than 4%. The average 30-day T-bill rate since 1990 looks like it is less than that. Does a long-maturity Treasury have some purpose in financial markets (collateral?) that is not served by T-bills? I don't think so, but maybe someone has more information.

    ReplyDelete
  10. Possible reasons:
    1. Risk management
    2. Cochrane's recent paper has a chapter about how maturity structure of treasuries influences the timing of inflation

    ReplyDelete
  11. Do you mean this paper?

    http://faculty.chicagobooth.edu/john.cochrane/research/Papers/understanding_policy.pdf

    ReplyDelete
  12. Good day! This post couldn't be written any better! Reading this post reminds me of my old room mate! He always kept talking about this.

    ReplyDelete
  13. I enjoy reading through your article post, I wanted to write a little comment to support you and wish you a good continuation All the best for all your blogging efforts. Appreciate the recommendation! Let me try it out. Keep working ,great job!

    ReplyDelete