The public policy blog of the American Enterprise Institute

Subscribe to the blog

Discussion: (20 comments)

  1. ….the stress and volatility in the U.S. financial system has returned to the pre-recessionary, pre-financial crisis levels of 2007.

    If my risk were backstopped, I wouldn’t be stressed out either.

    1. LOL!!!!!

      1. fallingman


  2. morganovich

    the vix is easily the most misused index in all of finance.

    the vix has NO predictive power except as a potential contrary indicator. it is at best coincident and really more of a lagging indicator.

    people buy puts like crazy when the market starts going down. that spikes the vix.

    it tends to be very low at market tops.

    track 2007 and this becomes clear. 2000 was the same way. by the time the vix is spiking, you are already in a serious down tape. using the vix to predict the future of the market is mostly like using your air bags to steer your car.

    the one exception may be extreme high values that tend to show enormous fear. when you have a whole market pricing in 90 vol, then you can assume that people are terrified.

    of course, making this bet in 2008 when the vix hit 89 in october 2008 was pretty questionable.

    the market kept dropping for 5 months and shed another 25-30%.

    the vix was dropping that whole time.

    one could also possibly try to use the vix as a contrary indicator as low levels show complacency and a lack of risk aversion that tends to come at the top of a bull run, but this is pretty useless as well.

    the vix can stay at low levels for long, long periods.

    it was as low as 11 in 2004 but was making fresh lows in the mid 9’s by 2007 (right before the first check back) so i would really not try to use this index that way either as the false positives are very high.

    all in all, the vix is just not a useful predictor of anyhting.

    1. the vix has NO predictive power except as a potential contrary indicator.

      Actually, the VIX is very much forward looking. It’s the implied vol for front month contracts. It is literally predicting the volatility of the underlying stocks for about the next 30 days.

      people buy puts like crazy when the market starts going down. that spikes the vix.

      When the market tumbles, expectations of further volatility rise, increasing the implied vol. That’s what spikes the VIX, not the put buying, per se. When the market drops, options market makers take their implied vol up long before put buyers show up. The index will reflect that.

      it tends to be very low at market tops.

      That market’s path is usually up a little every day for a while and then a sharp downward move (hence the skew in the puts). Obviously, during a period of calm when the market is drifting upward, implied vol tends to drop.

      1. morganovich


        i get what you are saying, but that front month contract implied vol is itself, not terribly predictive. it does not lead bad things happening, it jumps as the market reacts to them being reported.

        it’s coincident to trialing.

        we have played with models for years looking for forward r squared to equity prices etc and way to use vix as a hedge for a long portfolio.

        it’s all worthless. the vix just does not improve predictive ability.

        of interest, if you want to ahve some fun, look at the vix etf’s.

        the vxx is a hilarious snake pit.

        it rolls through the near and out month vix futures to maintain constant weighted duration. the contango will eat you alive over time.

        1. Morgan,

          You’re really twisting yourself into knots unnecessarily.

          The VIX is not a predictive model. It’s options market makers’ best guess about volatility over the next few weeks.

          An options market maker needs to price options. To do that he needs implied volatility (it’s an input to the pricing model). Since nobody knows what the future will hold, market participants have to guess. Your best clue when guessing is the volatility you’re observing currently.

          It’s not a commentary on whether the world is right or subsidized or Fed policy or whether or not we’re going to face a shit show three months from now. It’s just the market’s best guess. As is everything in forward looking markets since nobody is clairvoyant.

          the vxx is a hilarious snake pit

          There are some poorly constructed ETFs. I was talking about VIX options and futures.

          1. I will I could edit

            Your best clue when guessing is the volatility you’re observing currently and the supply and demand for options (both puts and calls. A call as a put and a put is a call. Remember?).

          2. morganovich


            i think you need to re-read what i wrote. i’m not twisting myself into anyhting, you seem to be misinterpreting what i said.

            my whole point is that the vix is not predictive. you seem to be agreeing with me.

            mark put it forward as a sign of financial stability and was attempting to use a low vix reading as a sign of a lack of “financial stress”.

            my whole point, with which you seem to be agreeing, is that the vix is not a useful predictor of financial stress or risk.

            implied vol on the cboe equity options does not have predictive power either in terms if guessing what the market will do or the economy or the financial system.

            we’ve really torn it apart looking for a correlation or some useful way to look at it (as we have with dozens and dozens of variables). i get your point about options pricing in risk, but the simple fact is that the aggregate of the vix does not give meaningful signals, even in the very short run like days or weeks.

            i’m a little confused by your responses here. you seem to be saying the same things i am (apart from maybe believing the vix has short term periodicity ability, i’m not quite sure what you think there) and yet also think you are disagreeing with me.

            am i missing some part of your argument or did you, for some reason, assume i was making a “the vix is predictive” argument?

            or are you just talking about the very short run? (where i still think the vix is all but worthless as a forward indicator)

          3. Morgan,

            I think you’re reading to much into “the VIX is not a predictive model” part of my comment. What I mean is that the VIX is an index and not a model. You can slice it, dice it, do whatever you want to it, but it’ll still be just the aggregate world opinion about very near-term risk. Whether or not you find it useful to make investments in stocks is an entirely separate and unrelated question.

            I am telling you that implied volatility is the market’s guess about future volatility. That is, implied volatility is a prediction. So, no, I do not agree that the VIX is not predictive. It is, in fact, the whole world’s prediction about risk in the next few weeks.

            mark put it forward as a sign of financial stability and was attempting to use a low vix reading as a sign of a lack of “financial stress”.

            That is correct. Depending on how you’re defining “stress”, Mark is correct. However, I think you need to step back and realize Mark isn’t using the VIX alone to make that point. If you disagree, then go buy some vol.

            Maybe you mean something else by “stress” and you need to let us in on your definition.

        2. In terms of possible uses of the VIX for you, here’s a suggestion:

          If you aim to run a portfolio with a certain delta, changes in the VIX tell you that your delta is changing even though your model hasn’t enough information yet to pick up the change in beta.

          For example: if you’re comfortable running your portfolio with a 500 delta and the VIX rises, your delta has also increased. Your betas have changed, but your model may not yet reflect that. The rising VIX tells you that you need to sell deltas. We can use “delta” more loosely here. If you’re hedged with a short position in a correlated asset, you can think of the unhedged exposure as “delta” and increasing your short leg or decreasing your long leg is akin to selling delta. And, of course, the reverse for a falling VIX.

          But all that has nothing at all to do with what I think Mark is saying.

    2. track 2007 and this becomes clear. 2000 was the same way. by the time the vix is spiking, you are already in a serious down tape.

      Correct. Implied vols change when observed vol changes. The observed changes in volatility change expectations of future volatility.

      Morgan, you’re misusing the VIX. All it illustrates is the market’s expectations of stock market volatility over the coming few weeks. It’s not meant to be used to predict volatility beyond that.

      Products based on the VIX are very useful for volatility trading. I can’t imagine the VIX is very useful for your style of long-term investing.

      1. morganovich


        i’m not misusing it. i’m not using it at all. i completely agree that it has no predictive value. i would even go further that you did and say it has almost no predictive value even in the very short run with regard to the market, gdp, or financial risk.

        that was my whole point.

        mark is tryign to use it as a measure of financial risk.

        my point is that you cannot use it that way.

        it’s a specialized index that means pretty much nothing unless you are trading options.

        1. Well, it doesn’t mean “nothing” unless you think that all market prices (all of which are similarly forward looking) mean nothing. Remember, options traders price risk. Things that will happen in the future.

          1. morganovich

            perhaps i was too semantically imprecise.

            clearly, any price means SOMEHTING.

            what i meant to convey is that the level of the vix means nothing useful in terms of predicting future stock prices, economic activity, or financial risk/stress and therefore that mark is misusing it as an indicator.

          2. The VIX represents the whole world’s best guess about the range in which the stock market will trade over the next few weeks.

            Using the implied vol of the option on stock XYZ, you can calculate the market’s expected trading range for stock XYZ over the next few weeks. Alternatively, if there is no option on the stock, you can use the stock’s beta and the VIX to calculate the market’s best guess for that stock’s trading range over the next few weeks.

            How accurate are those guesses? As accurate as any human guess about the future is. The benefit of the VIX is that it is the aggregate of many millions of opinions. The collective knowledge of the market.

            Mark is not using the VIX to predict economic activity. He is merely pointing out (I think) that market participants are pretty calm and not very worried about huge market moves. Put another way: all of the world’s collective information implies no large surprises within this time frame. That’s it. Nobody is worried a big bank will fail, no country will go bankrupt, no terrorist attacks will happen and no hurricane etc.hammer the oil fields in the Gulf ,etc – not before March 2013.

            That doesn’t mean that at some point in the future Moral Hazard won’t come back to bite us mightily in the ass once again. Your stronger argument is that these measures are warped by risk shifting.

          3. morganovich


            i get what you are saying. my point is that low levels of the vix are bad indicators of impending volatility if you look at them historically. it does not tend to warn you in advance of big downside moves, even in the very short run.

            it tends to be coincident.

          4. That’s true, Morgan. Because nobody is clairvoyant, nobody can predict surprises. That’s why they’re surprises.

            But, knowing what the entire world thinks about risk is still valuable information.

  3. morganovich


    these are not really different indexes.

    chicago and st louis use many of the same inputs in terms of rates and rate spreads and while chicago is broader, the additional rates and spreads and indexes it uses are all tightly correlated and related to the basic rates and spreads in the st louis index.

    all have been heavily adulterated/manipulates/affected by uber aggressive fed policy like zirp, twist, qe, etc.

    regardless of whether these indexes once evidenced predictive power, there is absolutely no reason to trust them now as they data upon which they are based does not mean what it used to.


    to argue that the fed printing money and buying assets until rates and spreads change is the same signal as rates established by market participants making assessments about risk/reward is pure cargo cult thinking.

  4. I disagree with the title and basic thrust of this article.

    Look more closely at all three. In each case, if you superimpose the graph for the last 3 or 6 months over the graph for 2007, the current values are a least a little higher.

    Very slowing getting close to the same values, yes. In the case of the VIX, you can say that the current value is roughly the average of 2005-2007. But the VIX bounces around; it just happens to be at the low end of its current range. It tends to have a well defined lower level of its range, but spikes upward. Compare the lower bound of its current range against that in 2005-2007 and it is clearly significantly higher now.

Comments are closed.

Sort By:

Refine Content:


Additional Keywords:

Refine Results

or to save searches.

Refine Content