Friday, June 7, 2019

Futures forecasts


Torsten Slok at DB updates this lovely graph on occasion. Here's what it means. Fed fund futures are essentially bets on where the Federal funds rate will be at various points in the future. Thus, you can read from the dashed lines the market's guess about where the federal funds rate will go -- assuming that the bets are priced to have an even chance of winning or losing.

Reading it that way, the market was systematically wrong from 2009 to 2016. It's something like springtime in Chicago -- this week, 40 degrees and raining. Next week, 75 and sunny. Week after week after week. In 2017, the market finally changed expectations to say, no, fed funds rates are not rising -- just in time to miss the actual rise in federal funds. Now, as in the blue line, market forecasts say there will be a big decline. But, as Torsten points out, why would the market be right today?

So what does this graph mean? Are market practitioners really that dumb? After all, there is a lot of money to be made here. When the graph is upward sloping -- as the entire yield curve was upward sloping from 2009-2016 -- and so long as rates don't rise, you can make a fortune borrowing short and lending long. And vice versa. In short, the difference between forward rate (right end of dashed lines) and spot rate (current fed funds rate) does a lousy job of forecasting where the spot rate will go -- and thus, mechanically, is a good signal of the extra return, positive or  (lately) negative you will get by holding long-term bonds.

The pattern is actually widespread and longstanding. Starting in the late 70s and early 1980s, Gene Fama wrote a series of papers on it, short term bonds, money markets, foreign exchange,  and (a favorite of mine) long term bonds (with Rob Bliss). Campbell and Shiller also found it in long term bonds, which Monika Piazzesi and I extended.  Piazzesi and Swanson show the pattern in federal funds futures.

There are three potential stories:

One: the market is dumb. People are dumb. Well, that's a nice story that can "explain" just about anything. But if you're so smart why are you not rich. Behavioral finance isn't that empty, and searches for common patterns in dumbness. However this graph is the opposite of the usual behavioral claim, extrapolative expectations, excess belief in momentum.  If there is a rejectable hypothesis in behavioral finance, this graph seems to reject it. (I welcome corrections to that statement in the comments.)

Two: there is a risk premium and it varies over time. For most of 2009-2015, the economy was depressed. People needed a good promised return, a coin more than 50/50, to hold the risk of long term bonds.  Once we exit the recession, the opposite pattern holds. Long term bonds should pay less than short term bonds, and maybe now the yield curve is finally waking up to that pattern. Naturally, I'm attracted to this story, but I admit it's a bit strained late in the upslope period.

Three: exit and entry to recessions is something like a rare event, a Poisson process. Such a process is like computer failure. The chance of the event is always the same, and does not increase as the length of time goes by. Recovery could happen any time. In a second paper that's what Piazzesi and I seemed to find for this pattern in bond markets. The market forecasts are right, in fact, and we just got 7 tails in a row. That is a speculative idea, and needs quantification.

Whatever the story, here is the fact: futures prices are not good forecasts (true-measure conditional means) of where interest rates are going. That fact is true not just of Fed Funds futures, but interest rates in general.

Update:

Torsten sends along an updated chart, going further back in history.


22 comments:

  1. More likely: the market pricing reflects the average of some probability distribution. At every date, the market price implies some probability in the right tail (positive growth/inflation surprise, higher rates) and the left tail (the opposite). When rates are already at a low level, the left tail of that distribution is constrained by the zero lower bound, so the average naturally shifts to the right, indicating "expectations" for higher rates. As rates rise, the left tail gets longer and "expectations" move lower.

    ReplyDelete
    Replies
    1. Good idea, but it does not explain why the market did not become less optimistic after its predictions didn't come true again and again and again.

      Delete
    2. The number line from 0.2 to infinity is larger than the number line 0 to 0.2, and so even if half the predictions are lower and half are higher than 0.2, the average is going to be biased upwards above 0.2 as an anomaly of the calculation rather than an indication of sentiment.

      Delete
  2. Timely post. Two comments.

    1. I am not aware of any futures markets that usefully (profitably) forecast rates or prices. None. Economists should perhaps avoid implying that markets exist where forecasts are accurate with any kind of useful accuracy. There are none.

    2. The bad forecasts in the wake of the 2007-2009 Great Recession are puzzling. The stylized facts of financial crises suggest that recoveries from such events are slow and long.

    In addition, the structural factors that drove high inflation expectations in the past are no longer. Labour union power has declined and supply markets have become increasingly global and competitive.

    Individuals like James Bullard can talk all they want about increasing inflation rates by cutting rates but they never address these structural changes and never acknowledge that the US central bank policy stance has been highly stimulative for most of the 21st century without any noticeable bump in measured inflation or inflation expectations.

    In passing, it occurs to me that economists who share Bullard's view also share a lot in common with President Trump. Trump and these economists want to fool the American public for their own good.

    ReplyDelete
  3. The expectation hypothesis is inconsistent with time-varying interest rates. So it is unsurprising that it fails... What I find surprising is how much people still obsess with it (academics included).

    ReplyDelete
  4. If I'm a CME technical trader, looking at Slok's graph, why wouldn't I set up a short spot/futures hedge? It seems like a contango to me.

    ReplyDelete
  5. to relate to explanation three: is It correct to say that the Fed Fund options also have large variance premium?

    ReplyDelete
  6. Luminaries of the economics profession have been predicting higher rates of interest and inflation for decades. Who could be better informed, more earnest or intelligent than, say, a Paul Volcker or a Martin Feldstein? Yet both have made siren calls to impending higher rates of inflation and interest rates for decades and decades--- such gloomy outlooks have been fodder for financial-media editorials also for decades.

    Judging from such commentaries and also the actions of Japan in regard to fiscal deficits and monetary policy, I have no doubt that Japan has been in hyperinflation since about 2007.

    Yet if I were to sit in a room with a Paul Volcker or Mark Feldstein and they again sermonized about the threat of impending inflation, I would again believe them.

    Perhaps the people who play the Futures markets also believe. ( Is it possible some people are merely hedging and don't believe?)

    ReplyDelete
  7. This is a reply to Pavel's above. I did not know how to post my reply directly below his comment. My reply goes as follows: I am sorry to say, but in my opinion your comment is not right. Market participants became less optimistic: and you can see that in the DB chart, since the slope of the dotted curves is leaning progressively to the right, after 2015. So what this graph means, asks the Post above? It si very easy to answer to that question, for a practitioner. The whole story goes as folows: up to 2015, market partipants put faith in the Fed's words, that ZIRP policy and subsequently QE policy would have produced sizable changes in the real economy stats. Market participants believed that we were all going "back to normal" after a few months of QE. From 2016 you can easly see that this belief is weakening. And (noticeable) the advent of Trump, and his "4% GDP growth policies", leaves market participants very skeptical from the beginning. Up to today: today we all see the data, we all hear Chairman Powell admitting (June 4, 2019) that QE did not work and is still necessary; and for what Trump policies are concerned, no Trump-effect is left anywhere, not in the macro data, not in the macro forecasts, nowhere (except the stock market, of course). DB chart above is commented with a question: why would the market be right today? Probably, it isn't: but these numbers describe only what people are thinking, so ... what's the point? What you are seeing in this chart it is a view, not statistical data. The market has been wrong, but what should be said of Jerome Powell, dec 2018, stating the "the economy is strong and in a good position"? If we take that as a point of reference, then things could get very much worse than the expectations tell us, actually. The question that really matters here, in my opinion, (from a practitioner point of view at least) is how can we have at the same time skeptical Fed Fund market participants and an euphoric stock market. Is the bond market too skeptical ... or is the stock market too easy to manipulate from the outside?

    ReplyDelete
  8. Incomplete spectra. The X axis goes over one recession cycle. There is the underlying monetary cycle which is observable in the long term ten year yields chart since 1980. Expectations analysis will give the results above if the expectations operator is done with partial spectral. We can see the expectations not noticing the long term, up front in 2008. Then sec stags became popular and the expectations begin looking long term at the end of this graph.

    But the expectations operator still misses a boundary condition, the barrier when millennials have no choice bu to exercise the 'right to coin'. The outcome of our MMT moment is coming into focus.

    ReplyDelete
  9. Maybe the Fed deliberately frustrates expectations.

    ReplyDelete
  10. It has been a few years since I've looked, but if you overlay the Fed's forecast for the same you will see that the Fed is far worse than the market at predicting their own behavior.

    ReplyDelete
  11. the futures are discounting the economy. The Fed is almost always wrong; so why correlate an efficient market with a bunch of academics with a flawed economic model

    ReplyDelete
  12. Given that Fed pretty much just follows the market, we have the market making an error understanding the market. The Fed is just an expression of the market.

    This means that the market has a hard time estimating future economic conditions, and understanding the relationship between economic conditions and inflation. This should not be surprising, as the current theories do not explain observations.

    ReplyDelete
  13. Suppose we view FFF not from a speculative angle (i.e. where will FFR lands at expiration) but rather a hedge, specifically an OTM put option with current FFR as the underlying (UL). Bear with me. Putting numbers on this - FFR = 2.4%. Purchasing Sept19 FFF @ 98.00 (2.0%) yields profit iff rates < 2.0% (put strike price). Value of contract is 0.4%. Holding UL constant, as time passes, value of the FFF contract/option must decrease from 0.4% due to "theta decay". BUT this is impossible without the strike price itself moving less OTM - which in turn *increases* the value of the contract, thereby tempering (but not eliminating) theta decay. Dueling factors.

    In other words, when the FFF contract (UL) is viewed as the option itself, a very interesting dynamic plays out. Like OTM options, I expect theta decay to accelerate as we approach expiration. Also with very high Fed uncertainty, there is a ton of IV baked in. I'm not precluding rate cuts, but both these factors lead me to believe FFF are overpriced and will drift toward 97.6 (2.4%).

    Admittedly, I may have taken my options analogy too far.

    Disclosure: I am short Jan20 FFF.

    ReplyDelete
  14. What would happen if the Federal Open Market Committe and the Board of Governors made a statement committing their policy for the next year, or perhaps indefinitely, to NO CHANGE in their IOER rate.
    There is a lot of good base money in the banking system. The bankers just need to digest all that reserve base - and and make some non-TBond investments and loans for rates above the IOER. No need to increase the reserve base for a decade or so, and we arrive at Milton Friedman's idea of simply freezing the monetary base.
    How would that work? Would it work better than "central planning"? At least the FOMC would continue to meet and discuss every 6 weeks or so, so the journalists who need "big momma" on guard are comforted.

    But the default expectation should be no change in IOER nor with anything else the FOMC messes with; await some event that calls for action or further debate.

    ReplyDelete
  15. Hi John,

    The 'forecasting' component of futures contracts is very small relative to the requirement that they be consistent. An FFFutures contract 12 months out must reflect the current yield curve 12 months out, otherwise there are arbitrage opportunities. This sort of yield curve 'roll down' trade is in fact a part of a lot of portfolios, but market structural reasons and significant risks make it not return much more than average. Overall, the price of futures have very little to do with forecasting as they almost always say the future will be exactly like the present with adjustments for inventory or carry costs.

    ReplyDelete
  16. It seems as though the future/forward market is always predicting a regression to the mean historical interest rate of 3%-ish. Over time that may be the statistically correct bet.

    Also, I think there may be a philosophical problem with saying a market prediction was "wrong" because reality didn't match the prediction exactly. The market bet at any given time is a mediation of the risks in all directions. For example, if there is a 50% chance of 6% rates, and a 50% chance of 0% rates, a forward rate of 3% would reflect a correct market appraisal of the risks. But there is a 100% chance that it will be "wrong."

    When current rates are at 0%, the forward rate will always be predicting a rise eventually. So it will always be "wrong" in retrospect if the rise doesn't happen.

    ReplyDelete
  17. That interest rates and many other economic variable are unpredictable is well established and reflects the very simple fact that the future is unpredictable. Interest rates essentially reflect all the information that is known at the time, but changes in interest rates reflect news (new information) that cannot be forecast. By the way this is why the expectations hypothesis of the term structure fails so miserably and why economists and policymakers shouldn't use it as a key ingredient in their models and policies (think forward guidance).

    ReplyDelete
  18. Torsten Slok's charts make for interesting commentary, but I am skeptical about the validity of the long-end of the dashed curves that he has shown in his charts, and on which his comments "the market was wrong" appear to be largely based upon.

    CME makes the following observation regarding futures prices for which there is no volume or no open interest:
    "Settlement prices on instruments without open interest or volume are provided for web users only and are not published on Market Data Platform (MDP). These prices are not based on market activity."
    https://www.cmegroup.com/trading/interest-rates/stir/30-day-federal-fund_quotes_settlements_futures.html

    CME futures are for the current month and out to the 36 month from the current month. A quick examination of the futures contracts indicate that beyond the current year, there is little activity and little to no open interest. The question that ought to be asked before drawing a conclusion from the charts presented above is: just what is the reliability of the data that Torsten Slok is using? If the pricing data is not reliable beyond say 12 months out, then much of Slok's chart and his conclusions become doubtful.

    ReplyDelete
  19. It is well known that interest rates and many other economic variables cannot be predicted. This is due to the fact the interest rates reflect essentially all of the information that is known at the time: Changes in interest rates reflect the news A(new information) that cannot be predicted. By the way this fact accounts for the massive failure of the expectations hypothesis of the term structure of interest rates. Economists and policymakers should understand this basic fact and not use the EH as a central part of their economic models or as a basis for monetary policy strategies such as forward guidance.

    ReplyDelete
  20. Would some of the players in the futures market not be hedging against a big swing in rates? If the current interest rate is low, and I've made a big bet that it will stay low, I would want a hedge against a sudden rise, so I'd buy futures based on higher rates (i.e., which would pay off if rates rose).

    Would such hedging/insurance lead to the graph we see?

    ReplyDelete

Comments are welcome. Keep it short, polite, and on topic.

Thanks to a few abusers I am now moderating comments. I welcome thoughtful disagreement. I will block comments with insulting or abusive language. I'm also blocking totally inane comments. Try to make some sense. I am much more likely to allow critical comments if you have the honesty and courage to use your real name.