• Last week's GDP was not expected. Markets still need to adapt.

    Before writing about the markets, I'd like to comment on the ZH article below, which has not attracted much attention:

    https://www.zerohedge.com/news/2019-...rithmic-trader

    This article is really interesting to better understanding how the markets have evolved in recent years.

    This article explains different types of Algos, but what is most interesting to me are not those algos that try to front run through execution speed and bid/ask manipulation. Instead, I found the Money Allocation (MA) algos to be the most probable cause of the coming volatility.

    Indeed, these algos try to allocate money dynamically on the basis of a process of big data analysis and machine learning that is mostly going on auto-pilot. This article states that although these MA algos' objective is to provide a trading edge, the competition is so fierce that such an edge is elusive and fund managers end up trying to level up their AUM (Assets Under Management) in order to be able to secure a 2% fee. Hence, these algos have become the backbone of a storytelling activity destined to meet sales objectives.

    But, interesting enough, the article states that the marginal cost for running algos is close to 0. Hence, large funds such as Fidelity and Blackrock have started offering 0.01% management fees. This means that trading is getting more and more concentrating into a few algo hands (Blackrock seems to be managing 6 Trillion $.) How do you make money with such a huge AUM.

    The answer is simple: you can only make money by following the long side of the market and by being an active part of market efficiency. If your fund is large enough, you can easily influence price discovery. I do not say that these large funds intentionally manipulate market prices, but their algos work both on data analysis and liquidity detection. They have basically replaced the Fed' QE activity by simply continuing to allocate funds to indexes buying.

    The result is the possibility of a huge disconnect between value and prices at the global market level. The author of the article states that hedge funds will soon disappear and that even large investment firms will be replaced by the likes of FB and GOOGL which have a direct connection to private investors.

    I do not buy that argument. There is a huge regulatory barrier of entry into financial advice activities and too many risks of lawsuits for these high-tech firms to venture into this field.

    What will probably happen is a total collapse of return on assets due to assets concentration. Algo trading based on nebulous unexplainable programs will disappear and traditional value based investment will come back in force. But this move will not occur without a collective understanding of the toxic aspect of algo trading activities. This understanding will be felt the hard way.

    The article explains very well how the adaptative aspect of some algos could affect the general market by pushing other algos out of their comfort programmed zone. It would be some sort of snowballing flash crash effect. But I believe that the greatest threat will be an external event that has not been programmed, such as the Kennedy assassination or 9/11. How would today's market react to such events? I would say probably a no-bid situation followed by a total break down/halting/reversing of trades.

    Last Thursday, we witnessed a very tiny tiny unexpected event: the GDP was higher than expected.

    This resulted in a two days selloff in Treasuries, gold, copper and some fixed assets (LQD, HYG)











    Even oil and materials were down.





    The US$ broke out on the news.



    But on the equities front, there was almost no movement: the S&P500 stayed high and



    the futures are still up in a straight line.





    However, something interesting happened in the past two days: the 20DMF stayed flat while the Cumulative Tick indicated buying activities.



    On the other hand, large investors in small caps were completely out of the market.



    The culprits are algos: they did not preview a strong GDP and hence had to balance portfolios out of fixed assets and back into equities. But these funds have no specific value based investment purpose: they just reallocate money and the best way to buy equities is to drip buy everything in such a manner as not to disturb prices. Of course, prices had nowhere else to go than up since money has been moving in the markets in the past two trading days, but it is obvious that large investors are out of that move.

    Conclusions:

    The S&P500 is now into its $2800 resistance level. If this level is cleared, I am sure that many algos are programmed to clear all the stops and push prices higher, with probably $3000 as a target.

    It is however interesting to keep a close look both at the 20DMF and at the Market Weighted Average Total Effective Volume pattern.

    The 20DMF is a 20 days oscillator that is based on the Large Effective Volume evolution by sectors.

    The Market weighted average Total Effective Volume pattern works differently. As a reminder, the TEV -Total Effective Volume- displays the pattern of the total volume that influenced the price on a minute by minute basis. Typically, index investing algos drip release money and try NOT to influence prices. But they do because they do influence the bid/ask process. In the current environment, the TEV measure might even be more relevant than the 20DMF simply because of the sheer size of index following.

    To calculate this indicator, I take about 1000 stocks and normalize their TEV values for the past 100 days between 0% and 100%. If the TEV is at its highest, it gets 100%. I then weigh each TEV normalized value by the capitalisation of each stock and average the results.

    This is what is shown in the Figure below. We can clearly see that when this indicator falls below its Pink average line, the price tends to fall further in consecutive days.

    We can also clearly see that at the right of the Figure, this indicator is presently weakening.



    Below is a measure of the extension of that signal above or below its average in percentage points. You will note that a fall below about -2.5% is where you need to be out of long positions.

    Comments 3 Comments
    1. 1bullseye's Avatar
      Very thoughtful and interesting Pascal! Thanks for this.
    1. stharp's Avatar
      Pascal,
      Where can I access the Market TEV Extension graphs?
      I closely follow the N100 & SP500 20d Large Player Strength graphs as trend indicators.
    1. Pascal's Avatar
      Quote Originally Posted by stharp View Post
      Pascal,
      Where can I access the Market TEV Extension graphs?
      I closely follow the N100 & SP500 20d Large Player Strength graphs as trend indicators.
      Follow the link below. You scroll down.

      http://www.effectivevolume.com/content.php?539-20DMF