Costs are eating my profits

2nd Oct '17 by Epitychia

18 comments 5821 reads

Most of the 90 -100 rank stocks are small cap. The average bid-ask spread on these with Hargreaves Landsdown seems to be roughly 5%. Currently i estimate my stock turnover is about 6 months ......a massive 10% drag on profits!

Anyone have suggestions?

I have thought of:

Selecting from 80-100 rank, maximum spread 2%
Selecting from rank 98-100 hoping the rank stays above 90 a long time
Selecting from stocks moving up to rank 90-92, hoping they go higher and then take a long time to fall below rank 90
Selecting only from value and quality, these appear to have a longer life, momentum profits seem to disappear after 6 months.

Help would be gratefully received.

Unlock with Email

Unlock with Google

Already have an account?
Login here

About the Author

Epitychia

Premium Member

Academic, teaching finance and investing. Over twenty years as a private investor.

18 comments

robin66

2nd Oct '17

I have found the actual spreads to generally be much tighter than the quoted spreads. Have you tried putting in dummy trades to find the actual prices?

HumourMe

2nd Oct '17

I filter among other things, to only select shares with narrow (fair) spreads. In general I try <=100 bps but in reality will accept <=200.

rjg

2nd Oct '17

I agree with robin66 - spreads on small caps aren't on average 5%, many are far tighter. That said, you are only really losing 10% of your profits through the spread if the share price doesn't move and you sell at the same quote as you purchased at. Given that you will be aiming for the price to go up, the actual impact on your profits will be considerably less (hopefully) and one could argue that you haven't really had profits taken from you since it was never possible to buy at the bid price anyway.

This example might help to clarify what I mean:

A share is quoted at 95/100p. You buy 1000 shares at 100p and it costs you £1000. If you immediately sold, then yes, you would lose 5% because of the spread, however you hold for six months by which time the price has gone up and is now quoted at 120/126p. You sell at 120p and make a profit (excluding dealing costs) of £200 or 20%.

Now, of course, you could argue that had you been able to buy at 95p and sell at 120p, you would have made £250 but on the day that you entered/exited the trades, this option wasn't available to you. Those prices, were to a certain extent, completely irrelevant to you - they were only relevant to the person on the other side of the deal.

For me, the most annoying bit about a wide spread is when I first purchase a share and it shows as a loss on my portfolio. Once the share rises past the breakeven point, I forget about it because the price is the price.

ISAallowance

2nd Oct '17

That seems to be quite a large turnover for a stockranks-based strategy. I started a stockranks ISA in April which is about 6 months ago, in that time of the initial 6 holdings only 1 has had to be replaced (SCISYS (LON:SSY), which fell to a stockrank < 50 after results). In that time you indicate your entire portfolio would have rotated.

Perhaps your selling criteria are too tight? I haven't really settled my sell criteria in stone yet, but I think I'll allow stockrank to drift in the absence of bad news as long as it stays > 50, and will certainly hold those that convert to highflyers with QM holding and V dropping.

Alternatively, perhaps you have some bias that is causing you to select high stockrank stocks that are more likely to drop than average?

robin66

2nd Oct '17

The other point which I missed earlier Epitychia is that you only "lose" the spread once so the "loss" on a 5% spread is just that;- 5%! and not 10%

ISAallowance

2nd Oct '17

If his average stock turnover is 6 months then he's losing the spread twice per year - 10% pa total.

Edited to add: In my opinion his core problem is the turnover, not the spread.

robin66

2nd Oct '17

Well you are right of course ISA on an annual basis. He didn't give a time period for his 10% loss and I assumed he meant 10% per trade. Re-reading the post I think he probably did mean per annum!

Logic

2nd Oct '17

Someone recently posted that they had a rather high turnover in a SR based portfolio. I suppose it may depend on the exact criteria of course.

That said, as some have pointed out, you may be able to get a better deal than the initial quote. In addition, you may be able to get a price closer to the lower price if you put in an order and wait to see if it gets filled (this assumes you pay no fees if the order does not go through, ofc). Alternatively, you could try to go for larger companies with decent stock ranks. Or you could have say SR>90 as a requirement to buying, and use say SR<80 for selling.

Flackwell

2nd Oct '17

You can also use limit orders in the hope volatility moves in your favour for the time necessary to effect your trade

Edward Croft

2nd Oct '17

I think this post is actually factually misleading.

Most of the 90 -100 rank stocks are small cap.

Not statistically true. Most are mid or large cap - here's a pie chart of the UK 90+ Ranked set by their size group.

59d2366550ab1Untitled_spreadsheet_-_Goog

The average bid-ask spread on these with Hargreaves Landsdown seems to be roughly 5%.

Again not true. The issue you are facing is that you are trying to trade sub £50m market cap stocks. These we classify as microcaps rather than small caps. The median spread of the 142 primary listed 90+ Ranked shares is actually 60 basis points. That's 0.6%. I guess the lesson is - be careful where you trade !

Size Group	Spread (%)	No. of Stocks
Micro	531	29
Small	198	41
Mid	20	48
Large	8	24

matylda

2nd Oct '17

Not sure if this is a joke?

pka

2nd Oct '17

Hi Ed,

The units of the 'Spread' column in the table in your post should be basis points (bps), not %.

peterg

3rd Oct '17

Edited to add: In my opinion his core problem is the turnover, not the spread.

I think that's too simplistic. It's the combination that's the problem. There are plenty of ways to skin a cat - but if you are going to adopt a system that ends up with frequent trades you need to keep the spreads small otherwise you could have to make 5-10% (which would not be a good average for trades lasting 6 months) on each trade to stand still. There a plenty of stocks out there trading at small fractions of 1% spread (even if they are not ones that get discussed here much) and if you want to trade short term you are likely to better concentrating on those with low spreads. If your preference is for small caps with large spreads you really need to be planning for the longer term and expecting your average holding period to be a fair bit longer than 6 months.

Peter

Epitychia

3rd Oct '17

Thank you for all the comments, they were very useful.

I looked up the 20 top ranking stocks on QVM, 2 were not available on Hargreaves Landsdown of the remaining 18 the average spread was 3.0%. I suggest this is less than normal because there are an unusual number of large cap. builders in this top group at present.

The main take away i wasn't fully aware of is the massive spread in the spread (no pun intended), from less than 0.1% on Persimmon to 7.7% on Synectics. Furthermore there does seem to be 2 distinct groups, those with a spread below 1.5% and those over 5% with few in between.

Going forward i will put in a spread limit of 2.5% before buying and there should still be enough candidates in the top ranks.

Again thanks for the help.

Finally.........it would be very useful to know the typical length stocks stay in the top decile for quality/value/ momentum and their cross overs ......i suspect there are large differences between them and therefore perhaps in the net profitability of these screens. Amino technology is a case in point, i owned 2 years ago then sold because the rank fell below 90 and then bought again recently.

pka

4th Oct '17

This has been an interesting discussion, but there hasn't been much maths informing it. Let's do some maths based on a few assumptions.

Let's assume that:
(1) The past performances of the various bands of the QVM StockRank over the last 4.5 years are a measure of their likely future performances, as indicated by their past performance graphs..
(2) However, we should bear in mind that the performance graphs IGNORE both transaction costs and reinvested dividends. I think it is reasonable to assume that the spreads of individual stocks have very little correlation with the dividends that they pay, so we will ignore dividends, but transaction costs and spreads are very important.
(3) The average spread of the stocks in the 90-100 band is 3%.
(4) Eliminating all stocks with spreads greater than 1% from our high StockRanks portfolio reduces the average performance of the highest ranked stocks from that of the 90-100 band to that of the 80-90 band, but also reduces the average transaction cost per rebalancing event by about 2%.
(5) The average transaction cost of a rebalancing event is 0.5% stamp duty plus 0.2% total stockbroker's commission for the buy and sell transactions plus the average spread, so it is about 3.7% in the case of the original set of stocks with the highest StockRanks, and about 1.7% for the highest set of stocks with spreads less than 1%.

Therefore each rebalancing event reduces the expected return by a factor of 0.963 in the case of the original set of stocks with the highest Stockranks, and by a factor of 0.983 in the case of the set of stocks with highest StockRanks and spreads less than 1%. These are equal to the reductions of the expected annual returns with annual rebalancing, i.e. reductions of 3.7% and 1.7% respectively, which is a difference of 2%.

However, with quarterly rebalancing, there are 4 rebalancing events each year, so the reduction of the expected annual return is by a factor of 0.963 to the power of 4, which is equal to 0.86, or a reduction of 14%, in the case of the original set of stocks with the highest StockRanks. And it is by a factor of 0.983 to the power of 4 in the case of the highest set of stocks with spreads less than 1%, which is equal to 0.934, or a reduction of 6.7% of the expected annual return, which is 7.3% LOWER than the reduction in the case of the original set of stocks with the highest StockRanks. However, reductions of 14% and 6.7% due to transaction costs from rebalancing quarterly are both HUGE hurdles to overcome, and will probably outweigh the expected benefits of having high StockRank stocks in one's portfolios. My conclusion is that it is NOT a good idea to rebalance quarterly.

According to these past performance graphs:
https://www.stockopedia.com/stockranks/performance/?exchanges=LSE%2CISD&field=_QVMScore_lo∩=10&period=4&buckets=deciles
https://www.stockopedia.com/stockranks/performance/?exchanges=LSE%2CISD&field=_QVMScore_lo∩=10&period=1&buckets=deciles

and the formula and calculator for Compound Average Growth Rate (CAGR) here:
http://www.investopedia.com/terms/c/cagr.asp
http://www.investopedia.com/calculator/cagr.aspx

I calculated the following CAGR values over the last 4.5 years for the 90-100 and 80-90 QVM StockRank bands rebalanced annually, ignoring transaction costs and dividends:

90-100 band rebalanced annually: CAGR = 18.7%
80-90 band rebalanced annually: CAGR = 14.8%

This represents is a difference of about 3.9% in the annual CAGR, which is bigger than the assumed reduction of 2% in annual transaction costs from restricting our portfolios to stocks with spreads less than 1%. So it might not be worth restricting our portfolios to spreads less than 1% if we rebalance annually, but it seems to be a marginal decision.

I also calculated the following CAGR values over the last 4.5 years for the 90-100 and 80-90 QVM StockRank bands rebalanced quarterly:

90-100 band rebalanced quarterly: CAGR = 20.4%
80-90 band rebalanced quarterly: CAGR = 14.9%

This represents is a difference of about 5.5% in the annual CAGR in the case of quarterly rebalancing, which is smaller than the assumed reduction of 7.3% in annual transaction costs from restricting our portfolios to stocks with spreads less than 1%. So it seems to be worth restricting our portfolios to spreads less than 1% if we rebalance quarterly. However, the transaction costs from rebalancing quarterly are likely to reduce expected annual returns by about 6.7% even in the case of restricting our portfolios to stocks with spreads less than 1%. So rebalancing quarterly does not seem to be a sensible policy.

steviej

4th Oct '17

I have a qtrly rebalanced portfolio that is based on 90+ StockRanks. I do not restrict based on spread or market cap.

It costs about 1% to do each rebalance.

pka

4th Oct '17

Hi Steviej, you wrote: "I have a qtrly rebalanced portfolio that is based on 90+ StockRanks. I do not restrict based on spread or market cap. It costs about 1% to do each rebalance."

I don't understand how that is possible. If you pay £10 stockbroker's commission for each transaction, that represents 0.1% if each stock holding is £10,000, and 1% if each stock holding is £1,000. As you need both a buy and sell transaction for a rebalance, the stockbroker's commission is likely to be at least 0.2%, unless the size of each of your holdings is much bigger than £10,000. If you add 0.5% for stamp duty on the buy transaction, your average spread must be 0.3%, which seems very small if you do not restrict based on spread or market cap.

steviej

4th Oct '17

£5 per trade, no stamp duty on AIM which is about quarter of buys, I don't buy and sell every share as some still qualify (or at least they always have for me), that would reduce the cost relative to the portfolio somewhat as well.

1% is average. Just did my rebalance Friday/Monday and the cost was just under 1%. Previous qtr was more like 1.5%.

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

Take a free trial

Facebook Twitter Youtube Instagram RSS

This site cannot substitute for professional investment advice or independent factual verification. To use it, you must accept our Terms of Use, Privacy and Disclaimer policies.

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

Take a free trial

Costs are eating my profits

Unlock the rest of this article with a 14 day trial

About the Author

Epitychia

18 comments

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

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