Sign in     Like us on Facebook Follow us on Twitter Watch us on YouTube

News and Blog

Join 5000 other sharemarket traders for regular blog updates!

Browse to a category

Blog Search

Even when they’re profitable every day, high-frequency traders aren’t making much money

Virtu Financial—one of the world’s largest computerized trading firms—made money every trading day last quarter. The problem is that it made less of it than in the past, as volatility in the financial markets has dried up in recent months. Big price swings are good for high-frequency trading strategies, as machines can swoop in and take advantage of market shifts.

While many high-frequency trading shops are secretive about their results and plans, Virtu is listed in New York, so its required updates provide a view into the state of the industry. The company’s profit from trading fell in just about every category last quarter, with net income from currencies and commodities taking the biggest hit, each declining some 30% versus the same quarter last year. The company’s share price fell by 8% in early trading.

More here – Quartz

Apparently Its All Over For Commodities

Commodities form an important part of my trading universe, in fact as a I par back my universe of instruments they form a more and more integral part of what I do. This pivotal role occurs for a few reasons ranging from ease of trading, price discovery and familiarity. The first trade I ever did was on a gold stock and the interest has stuck with me for decades.  You would therefore expect that I take an interest in the market so I had more than a passing interest when this article dropped into my news feed. There are a few points I want to dissect. However before doing that it has been my experience that there are two interesting phases in the life of any market. The first is when people start to tell me this time its different. Etched indelibly into my brain are the words of Irving Fisher who could be considered one of the worlds first celebrity economists who days before the 1929 crash uttered the immortal phrase stocks have reached a permanently high plateau. The second phase is that when someone tells you that a given instrument is stuffed beyond repair and will never go up again. Neither sentiment is reflective of either the cyclic nature of markets nor the psychology of traders.

Point 1 – Firms leaving the business.

This is an interesting point because it points to the number of prop firms leaving the business. However the number of retail investors exposed to commodities via ETF’s has grown dramatically so there has been a shift in the markets demographic away from wholesale to retail. The five largest commodity ETF’s managed almost $6 billion in assets.

Point 2 – Low Volatility

This point highlights one of my enormous bug bears it is the confusion between volatility and trend – the two are not the same and one doe not rely upon the other. I thought I would take a further look at this and just have a look at the distribution of volatility within the gold market. The first thing I did was simply look at the average 15 day volatility for a given number of years ranging from 15 years to the YTD and the results are shown below.

v1

Depending upon the look back period you can make a point that there has been a drop off over time in volatility . However volatility is relative concept and the current volatility in the gold market is sitting at 11.8% which is just below the average volatility for the YTD. Yet price has trended from around $1,000 to $1,350 and then back down to around $1,000. Looking at short term volatility tells us nothing about the trend. When looking at volatility I thought I might be missing something so I broke the look back period into five year blocks to get a sense of how it might have changed over time and the results are below.

v2

When volatility is broken into blocks you can see that over time volatility has increased and then tapered off a little. The so called halcyon days of two decades ago that every longs for actually had markedly lower volatility than recent times.

I should also point out that volatility in the crude oil market regularly spikes to beyond 60% so i am not sure where this missing volatility has ended up. Again it is probably the perennial confusion between trend and volatility.

Point 3 – Correlation

This point always interests me because people very rarely make it clear whether they are talking about price correlation or returns correlation. Most people when they talk about correlation talk about price without meaning to. The correct measure in such situations is the correlation between the returns across asset classes. True diversification is generated when you generate uncorrelated returns. Te first chart below looks at the daily performance correlation for gold, S&P 500 and crude oil.

p1

There is what appears to be an emerging positive correlation between gold and the S&P 500 and this surprised me a bit and I was suspicious that it was an artefact in the data so I generated a new series of data that looked back to the beginning of the century because my suspicion was that what I was seeing was actually the stagnation in gold and the rise of the US market post the GFC.

p2

Looking at the data over much longer term gives a clearer picture of what is actually happening. As US markets collapsed gold recovered and as US markets recovered gold suffered so to my eye the emerging correlation is somewhat of an artefact in the data. Much is implied in the article about how good commodities trading was in the past but it needs to be remembered that gold took 31 years to surpass its 1980 highs. That’s a long time between drinks if you are a long only trader as many commodities firms were. Commodities are the magic swing through mutli hundred dollar range tools that people think they are.

Point 4 – Leverage

An irrelevant point if you know what you are doing. Leverage has been a function of commodities markets from day one and is the staple of FX markets and they dont seem to have any problem coping.

Point 5- Liquidity –

I am not certain what the point is here since volume in the majority of commodity markets has increased dramatically over the past decade.

Point 6- Regulation

This is an old catch cry – if you dont know what you are doing blame the regulator. In some ways this is the same as football coaches who blame the umpire for their team being rubbish.

Point 7 – Its downright difficult

There is a particular sentence here I want to highlight –

For one, their idiosyncratic characteristics can make price forecasting practically impossible.

Price forecasting for all instruments is impossible. For those who need a quick refresher on how stupid this sort of thing is I give you Jon Boormans wonderful, regularly update guru predictions chart.

Predictions-2

Any attempt to predict price in any instrument is an exercise in delusional stupidity of the highest order.

The upshot of all of this is that the majority of things written about markets that have any sort of predictive narrative about the trajectory of a given market or markets is largely irrelevant and that includes this piece. The simple fact of all markets is that they are cyclical in both tone and the level of investor involvement. If I can defer momentarily to a local example. If you were to look at a comparison between housing and equities as an investment choice you would say that equities are dead. Yet funds continue to invest in them and prices continue to go up and down and some prices go up a lot.  The same is true for commodities and I doubt it will ever stop being true.

 

 

So What Does This Mean?

In my junk folder I have for the past upteen decades been getting random charts by a group called Chart of the Day. Surprisingly, I dont get them everyday – so the implication that you get a chart everyday that is interesting is perhaps a little bit of an oversell. This morning I go the following piece of wisdom –

c2

This chart as the title suggests looks at the S&P 500 PE ratio back to the turn of the century. Putting aside the obvious gaping methodological flaws such as the S&P500 was only started in 1957 I do always find these sorts of things interesting. Markets and their history should be a topic of investigation for every trader, simply because there is nothing new. Bubbles and crashes have been a feature of markets since they began and the driving force behind such things has always been the capriciousness of market participants. Curious as to what our own market looked like I dug up some data from the folks at Market Index and plotted the local PE ratio against the All Ords to see what I could see.

Capture

On the chart above I dropped a series of vertical lines – the three black ones denote a time when valuations according to the markets PE ratio could be considered extreme, the red one is the GFC. Pundits who look at valuation models work on the notion that markets or their component equities have a fair valuation and deviations from this point indicate that something is either overvalued or undervalued. Decisions are then made upon this interpretations. The first black line is easy to identify – its the 1987 crash.  The second one took me a little while to remember until I remembered the tail end of the 1991/2 recession combined with the banks nearly sending themselves under after property bit the dust.  The third black line is the tech wreck, The question when looking at any methodology is what value does it add to your decision making.  This is an important question since our decision making is bounded by the time we have to make the decision, the amount of information we have and our cognitive ability. None of these components can be infinite so our decision making is always somewhat half arsed. However, we need to add to this the notion of decision fatigue. It is estimated that during an average day we make anywhere between 20,000 and 25, 000 conscious and unconscious decisions and each of these decisions extracts a toll. Decision making is not a free ride, everything has a cost. Therefore efficiency of decision making is of paramount importance. If you have to force a decision then you are merely adding to your own mental loading without achieving anything.

As to whether the chart above tells me anything I dont already now about market extremes is doubtful As to whether it adds anything to my overall view of the world and approach to trading I am certain it doesn’t. But your mileage may vary.

Japan 1989

I am not one for historical comparisons because they are largely specious and mostly irrelvant but this one gives a nice bit of history that I lived through so consider it a nostalgia piece.

……The Nikkei stock index rose more than 900% in the 15 years before it finally topped. It was a frenzy powered by a belief that Japan Inc. was on its way to taking over nearly every major industry worldwide. The stock market bubble was further fueled by a massive real estate bubble at least twice the size of the one the US experienced in the 2000s. Tokyo alone became more valuable than all the land in the US.  In short, it was the product of a tsunami of monumental and concurrent events that are unlike anything present in the US today…..

It is hard to comprehend the apprehension that was rolling around markets and the business world with what was called the coming Japanese century. There was a belief that just 45 years after the end of the Second World War that Japan would rule the investment world. Even popular fiction writers such as Michael Crichton got in on the act with his novel Rising Sun. Japan was the flavour of the month, we were both fascinated and frightened.

 

Property Versus Shares

I was mucking around on the Valuer Generals site the other day searching for historical bits and pieces relating to my property when I noticed that the VG kept historical records on their estimation of the median house price in Melbourne. One of the things I have always found difficult in real estate is not the paperwork, the land rats, tenants, maintenance or the incredibly primitive way that houses are actually sold but rather the paucity of data that surrounds their instrument. Reliable and consistent data seems to be very hard to find and this was an issue I found when I was looking the the VG estimations – I couldn’t get them to tie in with other bits and pieces I found. However, only the VG site had any depth of historical information. Imagine trying to deal in a stock that had half a dozen conflicting prices from different sources, none of which you could actually deal in because prices are largely made up and then trying to find out what the price was five years ago only to get another half a dozen differing prices. It seems as if the real estate market is deliberately set up to be obscure and in some ways reminds me of an embryonic options market where those involved either didn’t understand their market very well or were being deliberately opaque.

Out of curiosity I downloaded the VG’s median house price data just to have a look at the trajectory. Because I am frequently bored I like to look at the history and structure of various markets – too few people are actually students of the markets they operate in. As such they miss out on a large number of free lessons that can short cut their process. No one makes original mistakes in their investing, everyone has made the same mistakes before you and the lessons from these mistakes can often be found in the data. Whilst mucking around with the data I remembered that the ASX often produces a comparison between the returns that are generated by various investment categories. I have always thought that these comparisons had a flaw in that they relied upon simple average returns, looking at averages is fraught with danger because they can be extremely misleading. Which is why managed funds constantly quote them.

As an extreme example consider the following investment scenario. I discover a magic fund with brilliant marketing material and on day one of year one I invest $100,000. In the first year the fund makes a return of 100% and I think I am a genius. In the second year the fund loses 50% and naturally I think the fund manager is an idiot but I am consoled by the fact that the year before I made 100%. When I present this scenario to people I ask them what the average rate of return is for those two years and most people answer correctly – it is 25% (100%-50%/2). I then ask how much have I made on my original $100,000 and I generally get an answer in the ball park of $100,000 x 25%pa. These guesses range from $125,000 to $150,000 as people try and do a compound interest calculation in their head. The truth is I have made zero – in the first year I doubled my money and in the second year I halved my money thereby returning me to my starting point. Yet my average return is 25%. This is why when looking at returns we have to be careful about using a long term average to generate an idea of how much we would have made. It is better to look at each individual piece of return data and assign a dollar value to it. This way you can build some form of equity curves which gives you a lot more information as to the trajectory of the value of your investment.

For a bit of fun I decided to take the data from the VG’s site and apply the returns from the All Ordinaries Total Return Index to their initial starting capital of $75,500 and see what the comparison between the two was. In effect I built an equity curve for the median house price and an identical investment into a surrogate ETF.

hp1

I have to admit I was a little surprised at the size of the differential because when you hear talk of comparisons between the two investment vehicles the impression you get is that the returns are quite close and that with the runaway bull market in housing that property has been the place to be for long term passive investing. Plotting data like this enables you to get a sense of the trajectory of price and to me two things are immediately apparent. Equities are more volatile in terms of a passive investment and this volatility is apparent in the impact of the GFC. Property moves a little like a truck, slow and steady whereas equities tend to throw themselves around a little. However, the shocks are not as severe as I thought they would be, 1987 is a blip that doesn’t appear and the tech wreck was a mild impediment. What did do the damage was the GFC and this is the problem with a simple buy and hold methodology.

To compensate for this volatility and to give a more real world flavour to our surrogate ETF I dropped the loss from the GFC to 10% from the historical 40.38% which is reflective of what actually happened when our macro filters kicked in and dropped us out of the market. The result of this simple fix is interesting.

hp2

The dramatically different result is simply a function of controlling runaway losses and not allowing them to have a detrimental impact upon your equity. Such a technique  is not rocket science but it does seem sufficiently difficult that it eludes all professional money managers.

Despite what the data says I am doubtful that it will convince die hard property advocates of anything – people with firm opinions are immune to data and it is hard to break the emotional bond that people have with actually owning something. And that is not really the purpose of the exercise as the advantages of equity investing over property investing are many , manifest and quite easy to elucidate. But is does serve as a salutatory lesson in what the differing mechanisms of presenting returns can tell us. It also tells us in no uncertain terms as to why  the worlds second richest individual is a share investor and not a property investor.

 

 

Damn….Beaten By The Kiwi’s Again

Click the link below the pic to be taken to some fascinating graphics.

20170318_WOC001

Source – The Economist

The Great Con Continues

Bloomberg recently breathlessly trumpeted the following regarding hedge fund behemoth Bridgewater –

Ray Dalio Makes Clients $4.9 Billion in 2016 as Paulson, Soros Falter

Their article was accompanied by the following table of returns –

returns

When you look at the table $4.9B is a lot of money in absolute terms and it is here that it all falls down since absolute terms are largely meaningless. So I reordered the data and turned the returns into relative or percentage returns and as expected things look a little different.

realtive

All of a sudden the returns are not that impressive and it wouldn’t be a posting without a chart with some sort of reference figure.

chart

As is typical of hedge funds they all under performed the S&P500 Total Return Index. I will be interested to see when the 2016 rankings for hedge fund salaries come out how much of investors gains have been swallowed up by the avarice of the funds owners.

General Advice Warning

The Trading Game Pty Ltd (ACN: 099 576 253) is an AFSL holder (Licence no: 468163). This information is correct at the time of publishing and may not be reproduced without formal permission. It is of a general nature and does not take into account your objectives, financial situation or needs. Before acting on any of the information you should consider its appropriateness, having regard to your own objectives, financial situation and needs.