The bad news: more capital required
The good news: adding more instruments to your portfolio is undoubtedly the most reliable way to improve your profitability.
The bad news: you need more money. Your cash will be shared out between different instruments, and each one needs to have the minimum level of capital for t hat product.
Let’s return to the example of trading two CFDs: Euro Stoxx, and 20 20
US 10-year bond. The respective minima are $6,500 and $6,000. 79 77
With the usual rule that you should use twice the minimum capital, the total capital required is: ($6,500 × 2) + ($6,000 × 2) = $25,000
This is considerably more than the $13,000 you’d need to trade only Euro Stoxx (twice $6,500), or $12,000 for US 10-year bonds (tw ice $6,000).
Don’t worry: when trading multiple instruments, you can safely ignore my recommendation to use double the minimum capital required . That rule made sense in the Starter System with a single instrument. Without it, if you lost money you could end up without enough capital to meet the minimum trade size requirement. But if you’re trading two instruments, and you lose some money, then you have another fall-back option: go back to trading a single instrument.
Suppose you are trading Euro Stoxx and US 10-year bonds with CFDs, but only using the minimum capital for each: $6,500 and $6,000 respectively, totaling $12,500. You are then unlucky with your first trade and have just $11,000 left. This isn’t enough to continue trading both instruments, but you can revert to trading a single instrument: Euro Stoxx. Of course, if you continue to lose money and end up with less than $6,500 you will have to stop trading completely (or switch to another instrument with lower minim um capital).
This approach is equally valid regardless of whether you start with three, four, 40, or 400 instruments. I discuss which order you should discard instruments later in the chapter.
More bad news: time commitment
Another disadvantage of trading multiple instruments is that it will take up more time. I automate my trading, so diversification is easy; the computer does the extra work with no complaints. But even if you are running your trading manually off a spreadsheet and some free internet data, it’s still relatively easy to trade more instruments without too much additional effort. With practice, and correctly set up spreadsheets, it should only take a few minutes to process each additional instrument – especially if you follow the time saving tips from chapter six.
The theory of diversification
Which instruments should you add?
I’m going to assume that you are already trading a Starter System with a single instrument, chosen using the criteria in chapter five: the cheapest instrument for which you can meet the minimum
capital requirement . Having done that, which instrument wil l come next?
The instrument you choose needs to satisfy the following criteria. Two of these should be familiar from the Starter System, but the t hird is new:
The additional instrument must be cheap enough to trade (under my 20 20
speed limit: 79 78 maximum risk-adjusted cost of 0.08, and ide ally lower).
You must be able to meet the minimum capital requirements for both the first and the second instrument (though you do not need double the minim um capital).
The choice of instrument should maximise diversification . You get more diversification by adding instruments that are as different as possible from the instruments you a lready have.
Back in chapter one I discussed the various different asset classes that instruments belong to: FX, shares, bonds and so on.
Adding a new asset class is the most diversifying thing you can do. If you’re already trading Euro Stoxx CFDs, then adding another equity index like S&P 500 won’t provide much diversification benefit – the instruments are just too similar.
Instead, you should choose an instrument from another asset class.
But which asset class sho uld you add?
Table 3 8 will help.
Table 38: Approximate correlations of trading strategy returns STIR: Short-term interest rates. Vol: Volatility. FX: Foreign exchange. Comm.: Commodities. Crypto.: Cr yptocurrency Correlation values are long-run averages derived from back-tes ted results.
Table 38 shows the correlation between Starter System trading strategies for instruments from different asset classes.
Correlation is a measure of diversification. Lower values for 20 20
correlation mean more diver sification. 79 79
To use this table, start by selecting the row which relates to the instrument you currently have in your Starter System. Now, going along the row look for the lowest correlation values. These are the asset classes which you should consider adding next.
Once you’ve decided the possible asset class(es) the other two criteria come into play: cost and minimum account size. You need to pick the cheapest instrument you can, given the capital you hav e available.
What if you want to add additional instruments? For example, suppose you’ve already got the Euro Stoxx and US 10-year bonds in your trading account, but still have surplus capital? You just repeat the process as before, although now when you examine table 38 you would look at both the equities and bonds rows. You’d probably not want to add STIR (correlation of 0.4 with bonds), or volatility (correlation of 0.5 with equities), but anything else would d o just fine.
You can then proceed to add instruments from other asset classes until you run out of money to allocate or run out of asset clas 20 20
ses to add. ¹70 70
Diversification for CFD traders
In theory CFD traders have access to a wide variety of asset classes, although not all instruments will be cheap enough to trade or have achievable levels of mini mum capital.
Table 39: Indicative costs and minimum capital for CFD traders Minimum capital has been rounded up. Cost calculations using appendix B, and minimum capital calculations using formula 21 .
Where possible, the cheapest (lowest risk-adjusted cost) and most accessible (lowest minimum capital) instrument are shown for each asset class. Only instruments with trading costs less than 0.0 8
are shown.
Table 39 has my recommended menu of options for CFD traders. Like all menus, there are choices for the well-heeled, but also some cheaper dishes. Where possible, I include two instruments for each asset class. For wealthy traders, I’ve shown the cheapest instrument in the relevant asset class. There is also a more democratic choice: the instrument with the smallest minimum capital that meets my speed limit (risk-adjusted cost of 0.08 or less per year). Not all asset classes are included, and not all asset classes have two options. This is because I’ve excluded all instruments that cost more than my speed limit.
As usual, the figures in these tables reflect current market conditions, and you should recalculate them using the methods in appendix B before you make your ow n decisions.
For the CFD examples in this chapter, we began with the Euro Stoxx used in chapter six which is in the third row (equities) of table 38. Looking along the equity row, the volatility asset class (e.g., VIX) comes in at 0.5, most other assets measure 0.1
and commodities have a correlation estimate of 0. Ideally then you’d choose a commodity as your next instrument; failing that, something from another asset class – except volatility – wo uld suffice.
Looking at table 39, possible commodities for CFD traders include crude oil, but that has a relatively high minimum capital:
$15,000. If that is too rich for your taste then you should opt for another asset class, as long as it isn’t volatility. The US
10-year bond has a minimum capital of just $6,000, so that would be a good choice for your second instrument if you haven’t got the capital for crude oil. With sufficient capital you could trade all three: Euro Stoxx (equities), US 10-year bonds (bonds), and crude oil (c ommodities).
Based on my calculations, there aren’t any further asset classes 20
which a CFD tra
der should consi der adding. ¹70¹
Diversification for spread betters
The Starter System begins spread betting with a commodity: gold.
Looking across the sixth row (for commodities) of table 38, all asset classes apart from FX have a correlat ion of zero.
Table 40: Indicative costs and minimum capital for spread betters Minimum capital has been rounded up. Cost calculations using appendix B, and minimum capital calculations using formula 21 .
Where possible the cheapest (lowest risk-adjusted cost) and most accessible (lowest minimum capital) instrument are shown for each asset class. Only instruments with trading cost less than 0.0 8
are shown.
From table 40, the Euro Stoxx spread bet would be an excellent choice for a second asset: equities and commodities have a correlation of zero in table 38, and it has a relatively low minimum capital (£1,400). With more capital you might consider adding the Nasdaq equity index, which is amongst the cheapest instruments shown in table 40.
What if we had the cash to add a third instrument?
With a commodity and an equity, we look again at table 38. There are no strikingly high correlations between commodities, equities, and other asset classes, with the exception of the 0.5
correlation between equities and volatility, and the 0.2
correlation of FX and commodities. So, we have a strong preference to avoid volatility and a slight prejudice against FX.
I would add a bond – gilts if you have sufficient capital, or US
10-year bonds if your account value can’t stre tch to that.
For a fourth spread bet, I’d add an FX rate such as EURGB P or NZDUSD.
Diversification for futures traders
I recommended in chapter six that futures traders use corn as their first instrument. Looking across the sixth row (for commodities) of table 38, all asset classes apart from FX have a correlat ion of zero.
Table 41: Indicative costs and minimum capital for futures traders
Minimum capital has been rounded up. Cost calculations using appendix B, and minimum capital calculations using formula 21 .
The cheapest (lowest risk-adjusted cost) and most accessible (lowest minimum capital) instrument are shown for each asset class.
If you are a futures trader who is short of capital then, after examining table 41, you should first add European volatility.
Volatility is highly correlated with equities, so for the next instrument I suggest US 5-year bonds. Bonds and short-term interest rate futures (STIR) have a high correlation, so the fourth asset class should be FX. MXPUSD is a good choice. Then I would include a STIR market, and only Eurodollar meets my speed limit. Finally, you should add an equity market; Euro Stoxx has the lowest minimum capital requirement.
Wealthier futures traders can focus on minimising trading costs.
The Nasdaq equity future is ultra-cheap and would be the second instrument I would add after corn. Equities and volatility are correlated, so we have to look elsewhere for our third instrument. EURGBP is an ultra-cheap FX instrument that is a worthy addition to the portfolio. A cheap bond future like UK 10-year is next. Bonds and STIR are highly correlated so, for the next instrument, I would suggest an unorthodox choice: Bitcoin (unless, like me, you are violently opposed to trading cryptocurrencies on principal).
There are only two asset classes left, and the correlation of Bonds and STIR (0.4) is lower than the correlation of equities and volatility (0.5). So, I would choose a STIR market next, and only Eurodollar meets the speed limit. Finally, I would add a volatility market – the VIX is t he cheapest.
Diversification for margin traders
How can margin traders diversify across different asset classes, when they can only trade shares? Trading in both Facebook and Apple provides precious little diversification, as they are both US technology stocks. However, there is the option of buying exchange traded funds (ETFs), and there are ETFs available for many different asset classes including bonds, currencies, volatility, and commodities.
There is a varied and ever-changing number of ETFs available on the market. For this reason, I haven’t included specific examples in t his chapter.
Diversification for spot FX traders
Diversification is particularly difficult for spot FX traders, as they can only trade a single asse t class: FX.
Table 42: Indicative costs and minimum capital for spot FX
traders
Minimum capital has been rounded up. Cost calculations using appendix B, and minimum capital calculations using formula 21 .
Only instruments with trading cost less than 0.0 8 are shown.
Assuming you begin with a developed market pairing like AUDUSD, the most diversifying instrument to add would be an emerging market currency, like MXPUSD. However, wide trading spreads mean that these instruments are often too expensive to trade; MXPUSD
has a risk-adjusted trading cost of 0.15. This is well above my speed li mit of 0.08.
You will have to settle for adding other developed market pairs; a selection is shown in table 42. For obvious reasons EURGBP is highly correlated with GBPUSD and EURUSD. Because of their close economic ties, AUDUSD and JPYUSD are also fairly correlated. Bear these results in mind when choosing your currencies. For example, if you begin with AUDUSD then you should avoid adding JPYUSD
next.
If your instruments come from a single asset class, the instrument diversification multiplier (IDM) will be lower. I will explain how you can adjust the IDM appropriately later in the chapter.
Mix and match
There is no reason why you can’t trade different instruments with different leveraged products, assuming there are no regulatory barriers. So, for example, a UK-based spot FX trader trading AUDUSD could add a gold spread bet to their account, getting access to c ommodities.
Further div ersification
Once you have one instrument in each asset class then you can continue to diversify within asset classes; as long as you have sufficient funds, and don’t run into issues with the speed limit of risk-adjusted costs below 0.08. Here are the priorities for further diversification. First of all, for commodities: Instruments from a differe nt grouping: The commodities groups are: Agricultural, Metals and Energies Next: instruments from differen t subgroups: Agricultural subgroups: Grains, Softs, Meats Metal subgroups: Pr ecious, Base
Energy subgroups: Crude oil and products, Natural Gas Finally: instruments within differen t subgroups: Grains: Wheat, Corn, Soybean (also Meal and Oils), Oats , Rough rice
Softs: Cotton, Orange Juice, Cocoa, Lumber, S ugar, Coffee Meats: Live Cattle, Feeder Cattl e, Lean Hogs Precious: Gold, Silver, Platinu m, Palladium Base me tals: Copper
Oil and products: WTI Crude, Brent Crude, Gasoline, Ethanol, Heating oil
Natural Gas: US Natural Gas (Henry Hub), UK Natural Gas Now for the other asset classes: FX, equities, volatility, bonds, and STIR; your priorities should be as follows: Firstly, add instruments from countries at different stages of economic development:
Emer ging markets
Devel oped markets
Next, add instruments from different geographi cal regions: Emerging markets: Central and South America, EMEA (Europe, Middle East and A frica), Asia
Developed markets: North America, Europe, Asia (includ ing Oceania)
Instruments from differe nt countries
Finally, for individual eq uities only: Equities from different indu stry sectors Also, for bonds (or bond in dices) only: Bonds with different cr edit ratings
Bonds with different maturi ty profiles Which instruments should you remove if you lose money?
Earlier I noted that if you lose money, you should remove instruments from your trading account. But which instrument should you r emove first?
Use this simple method: remove the last instrument you added when you were building your initial trading strategy. For traders who started with three or more instruments, if you have already removed one instrument an
d things continue to go badly, then remove the penultimate instrument you initially added, and so on.
This is a ‘last in, first out’ policy.
What if you become profitable, and your account value reaches the point where an instrument was previously removed? Then you can start adding instruments again, beginning with the last instrument you removed: ‘last out, fir st back in’.
Sharing out capital between markets
If you trade S&P 500 futures, which require capital of $178,000, then there is no benefit in adding US 10-year bond CFDs with capital of $6,000. Around 97% of your returns will still be coming from the S&P 500, with just 3% from US 10-year bonds. This is diversification in name only. You are effectively trading a single instrument.
For maximum diversification each instrument in your trading account should have exactly the same capital allocation . This isn’t always possible, but you should try and pick instruments to get as close as possible to equally weighted capital. Then, if you have the extra funds, you can get to an equal split by adding just a little more cash to the account.
For the example I’ve used throughout this chapter, we trade both Euro Stoxx and US 10-year bonds using CFDs. The minimum capital 20
for each is $6,500 and $6,000 respectively. This is a split ¹70²
of 52% and 48%, very close to even. But, if you can manage to add another $500 in your trading account for US 10-year bond trading, then you will have exactly $6,500 in each instrument.
With more than one instrument in each asset class, you should first share out the capital between asset classes, and then split 20
it between the instruments in each a sset class. ¹70³
Calculating the IDM
The instrument diversification multiplier (IDM) is the factor we use to calculate the target risk on individual instruments: we multiply the account level risk by the IDM. It exactly compensates for the reduction in account level expected risk which we see when we trade several instruments. This ensures we hit our risk target, and it converts higher Sharpe ratios into hig her returns.
Calculating the theoretically correct instrument diversification 20 20
Leveraged Trading: A professional approach to trading FX, stocks on margin, CFDs, spread bets and futures for all traders Page 16