Value & Momentum - The Strategy That's Crushed The Post-Brexit Market Rally

17th Aug '16 by Ben Hobson

48 comments 21484 reads

UK stock market indices have broken out to new highs this month, which is an impressive change of fortune from the Brexit-induced price collapse we saw in June. Despite worries about the economic outlook, it seems the market is taking an optimistic view (helped by some new economic stimulus from the Bank of England). Naturally there’s been a lot of speculation about which stocks may or may not do well in these conditions. But a dig into the data shows that it’s been Value and Momentum - cheaply priced stocks on a price and earnings uptrend - that have been the biggest recent winners.

Market Performance

In terms of numbers, the FTSE All Share has risen by 16.3% from a low in late June - pushing it to a gain of 9.3% so far this year. That trend has been broadly similar across the FTSE 100, 250, SmallCap and AIM All-Share indices.

Six-month performance of the FTSE All Share (black), FTSE 100 (orange), FTSE 250 (green), FTSE SmallCap (blue) and AIM All-Share (turquoise).

Financial commentators have given a lot of credit for this market uplift to the stellar gains in stocks that are somewhat protected from the domestic economy. In particular, they’ve pointed to commodity sectors like mining and oil & gas as some of the best. There’s no doubt that there have been some decent gains here, with the likes of BP, Shell, Rio Tinto and BHP Billiton all rising sharply on the Brexit vote as investors looked for safe havens. The same was noticeable in small-cap resource stocks where there were sharp rises in companies like Ithaca Energy and Ferrexpo.

Factors that drove the Brexit vote rally

Nearly two months have passed since the EU vote, and some early patterns are emerging on where the strongest return drivers have been. Stockopedia members will know that we score and rank every stock in the market based on the strength of its Value (whether it’s cheap), it’s Quality (whether it’s financially strong) and its Momentum (whether it’s price and earnings trends are improving). You can read more about how the StockRanks work here. What we’ve seen in recent weeks is that stocks with the greatest exposure to Value and Momentum have delivered the strongest gains.

While that top decile…

Unlock with Email

Unlock with Google

Already have an account?
Login here

Rusty2

22nd Aug '16

Something else I have just noticed if you do companies >1m then the 0-5 and 95-100 have about the same performance +103%. I also notice that the larger companies M scores have not done so well, so if you could just pick companies under £50m these would have performed very well but I can't do it on here. So in smaller companies the dogs (0-5) seem to do as well as the 95-100 M scoring companies which is interesting.

Depends on the size of the companies, >50m it's the other way around. In fact almost 10% difference. That's also the case for semi-annual.

Hot Socks

23rd Aug '16

Its fascinating isn't it if you have a play with the graph tool. there's a period where the bottom ranked momentum stocks seem to out perform just about anything else. I'm just looking at Value with companies >350m rebalanced annually, split into quintiles. Ranks 20 to 40 beat all others.

it does seem that in general higher rankings tend towards better results as time progresses, but the clearest picture of the rankings really working is with M, QM, VM or QVM with quarterly rebalancing >£10m. It can get quite patchy elsewhere. I wonder if M is just an easier concept to "bottle" as a statistical test. Like you say it would be really interesting to break down that performance into constituent shares and see what is happening at a more granular level.

Edward Croft

@hotstocks - it gets patchy at higher market caps because the lower ranked buckets end up having so few stocks in them - to test more effectively you'd want to use real quintiles with exact numbers in each bucket (not buckets dissected based on the stockrank ranges).

Similarly - for the lowest market caps they can get skewed by anomalous price moves in very low priced / low liquidity stocks.

We chose a £10m cutoff as default as that's a sensible cutoff liquidity wise for most private investors, and it covers the top 50% of stocks listed in the market. Last time I looked, 50% of stocks are priced below £10m on LSE - amazing really.

Yes - in general the research shows that Momentum works best over shorter time periods (as fashion lasts a season... 3 months to 1 year), while value works best over 1 to 2 year time periods (it takes time for value to out).

What's great about Value & Momentum is that one can get the best of both worlds. There's a lot of factor rotation in the market... sometimes Value is in favour, sometimes Momentum... they can zig/zag and their lower correlation with each other actually helps risk-adjusted returns.

There's bags of research on this... here's a write up I did on one of my favourite papers - Value & Momentum Everywhere.

Blissgull

Ed, why are the stockranks performance charts now telling a different story? It now looks as if the Brexit momentum rally which had showed the top momentum ranks as doubling never happened.

Ed, the momentum figures have totally changed since yesterday,what is going on?

If you look at my post 29 above, these are totally different now.

Yes I spotted that and know what happened.

We have a vast prices database so we only update the latest OHLCV price set for each day. When stocks split, a split event is reported and we re-import the entire adjusted price history for that security since flotation. There was one stock whose latest reverse split went unreported by Thomson Reuters until yesterday. So its price history was only reimported yesterday. We have another process that re-imports all stock price histories on an infrequent schedule so these anomalies are swept up, but this was a very small stock so was on a slower schedule.

If you check under the charts we do make this process clear.

6. Splits: We are vigilant about finding splits and consolidations - but occasionally they can be missed by our data provider and skew results. If you do see any anomalous charts please contact us using the green button and we will fix.

Sorry if you've been discussing a data anomaly... but that's the nature of data. We work very hard to keep our systems clean of bias and historically accurate. There can be occasional edits to pricing data when we do receive new information about corporate actions. If anyone has worked in a data research firm or systematic hedge fund they will know all about the work that goes into keeping data bias free.... there's a lot of work involved.

I appreciate that it's a nuisance, but I'm sure you'd rather there was a team working hard to ensure accuracy than the alternative. Most databases don't have thousands of investors poring over them every day and helping spot anomalies... it's all you guys who keep us on our toes and make sure we make the database as accurate as possible. Ultimately everyone benefits.

pka

What I find interesting about the performance charts for all the StockRanks, not just the VM one, is that there are two periods of about one year when the deciles fanned out, and two other periods of about one year when the deciles were all roughly in parallel with each other. In the periods when the deciles fanned out, stocks with high StockRanks performed very well relative to the market as a whole. However, in those periods when the deciles stayed roughly in parallel with each other, stocks with high StockRanks seemed to me to perform not much better than the market as a whole. Therefore if one uses StockRanks to select a portfolio of stocks, I think one would need to have the confidence to stick with that strategy even after a period of mediocre performance relative to the market as a whole.

Thanks Edward for looking into this. However this may explain the surge post Brexit in the 90-100 M shares but I don't think it explains the 0-5 M figures changing radically, some of these were 1-2 years ago.

herbie - send me the link to the test in question and I'll look into it.

Therefore if one uses StockRanks to select a portfolio of stocks, I think one would need to have the confidence to stick with that strategy even after a period of mediocre performance relative to the market as a whole.

This is the whole point. Strategies don't 'work' all the time. The StockRanks are a form of factor investing. The higher returns to factor investing are effectively a form of risk premium. The "Value Premium" or the "Momentum Premium" can be thought of as compensations for risk... they are like insurance premiums. They have a tendency to pay off (collect premiums) because there are associated risks of underperformance.

Some people don't want to own Value stocks because they don't want to own weak companies that are at risk in times of tougher credit. Some people don't want to own momentum stocks as they fear reversals or they fear trends don't persist. Whether the risks are behavioural or economic, they are real. Momentum and Value can both underperform for significant periods of time.

If things worked forever then everyone would do it and there would be no outperformance. As equity investors we have to welcome temporary strategy underperformance (or market neutral performance) as we know everyone else will give up and go and chase the latest hot story (Sirius Minerals)... I'd rather stick to what I know has worked for decades/centuries.

Not sure I can now as it has changed. Maybe I was mistaken.

I'm looking at this screen now: http://www.stockopedia.com/stockranks/performance/?exchanges=LSE%2CISD&field=_MomentumScore_lo?=1&period=1&buckets=vigintiles

I'm struggling to understand how the >£1m companies 0-5 M are doing so well and far better than companies >£10m, is it possible to produce stats for companies >£1m <£10m?

In future would it be possible to show companies less than as well as greater than £m? Because if you are interested in smaller companies you are not interested in the large ones.

I'm struggling to understand how the >£1m companies 0-5 M are doing so well and far better than companies >£10m, is it possible to produce stats for companies >£1m <£10m?

If you look at Vigintiles this is par for the course. Here's the £1m-10m mktcap set for the Momentum Rank as requested. You can see the "0-5" ranked set outperforming... let's investigate.

57bc8c1b1791fStockRanks_QVM_Performance_

Why does the 0-5 set run so far ahead? It's simple. In late 2013 there was a huge run in this set... but there were only 20 stocks in that bucket. 3 stocks rocketed more than 200%.

One was for Plethora solutions which flew about 470% in the quarter.

57bc8c75ed2e3PULSE_ENERGY_Share_Chart___

Sadly it's now delisted.

Another one was CPP Group... which flew more than 350% that quarter.

So when you narrow your search to just a few stocks (vigintiles)... and pin it at the very speculative nano-cap end of the market (where stocks move a lot on rumour) you're going to see some really anomalous moves.

Personally I wouldn't read into it at all... I think you are trying to read signal from noise. You could narrow the searches down to centiles and really have some fun !

Edward, thanks very much for showing that information, it's very interesting, having seen the companies I think you are right about reading too much into them, I will stick to a more balanced QVM stocks over £10m.

Ken Lowes

24th Aug '16

Is it me or have I missed something momentum means stocks going up or maybe down. If the market overall is going up then wouldn't the best momentum stocks automatically give the best return. Conversely if you have a portfolio based upon this theory then if the market turns against you won't the pain suffered be greater. Furthermore it isn't true that what goes down quickly comes back quickly or even at all. If one is looking at a ranking system what happened to the value of shares that disappeared from the rank, even if the rank remained constantly monitoring all shares it is still obviously going to do well because it is based upon performance, not quality. Another question when do you jump aboard these rocket ships when it is obvious or when it is more than obvious which could be at the same time as it runs out of fuel. If it runs out of fuel when do you sell? Does all this seam very basic or is it just me!

PJ0077

Ken

All price momentum studies focus on relative momentum, thereby neutralising whether the market is going up, down or sideways.

Academic studies shine brightly on momentum investing, even though it would appear that such a style of investing refutes the Efficient Markets Hypothesis in it's weakest form.

A great paper to read is "Fact, Fiction & Momentum Investing" by AQR which looks at the empirical evidence. It can be found here:

http://www.dorseywright.com/sites/default/files/SS..

All price momentum studies focus on relative momentum, thereby neutralising whether the market is going up, down or sideways.

Actually that's not strictly true ! There are lots of absolute momentum studies... the best book on the topic is Gary Antonnacci's Dual Momentum - can't recommend it more... but it's much more of a macro book. It does have some fantastic sections summarising the behavioural reasons for momentum and a great case for using absolute momentum as a timing measure.

I stand corrected, Ed!

Will read Dual Momentum with interest this weekend!

25th Aug '16

Thanks PJ
I will read the paper and come back to you.
Regards
Ken

Value & Momentum - the strategy that's crushed the post-Brexit market rally

Market Performance

Factors that drove the Brexit vote rally

Unlock the rest of this article with a 14 day trial

About the Author

Ben Hobson

48 comments

Unlock the metrics for all these stocks and more...

Unlock the metrics for all these stocks and more...