'Purest' global bond index fund

dunstonh

However it does seem that while investing solely in a global index equity fund is not irrational, investing in a hedged global bond index is less sound.

Indeed.   In theory bonds should be relatively low in risk.  However, currency fluctuations can create more volatility than the bond itself.   If both sterling and the home currency of the bond go in the right direction it can make the returns look much better. However, if they both go the wrong way then you get greater losses.  And as the point is to reduce the volatility, the global bond does not achieve that objective.    If the running yield was good (relative to the risk) and exchange rates were expected to move in the right direction then they may be attractive.  If they are all in the wrong place, then you want to avoid.  And that is pretty much where we are at the moment.

There comes a point where they are not going to reduce volatility and you would have to either reduce your equity allocation to counter that or alternatively use cash and gilts despite their low return potential and increase the equity content a little to compensate. i.e. instead of having cash, gilts, bonds and equity, you drop the bonds and have more in cash/gilts and more in equities.

bowlhead99

aroominyork said:

There is no one-size-fits-all solution but it seems good enough to use Jupiter Strategic Bond, which I consider a well-managed and relatively conservative fund, as my point of reference.

I generally don't think that any one fund in a sea of rival funds is a great benchmark because they will all go about their allocations in a different way, having different views on the market and tactics to employ to achieve the objective. So one year, well done you might have beaten Jupiter but perhaps Jupiter was in the bottom quartile. A strategic bond sector average would be more useful if you are not going to take a view  for yourself yourself on how to position your holdings and would in practice let a bond manager allocate all the money you give him with free rein to pick 'any kind of bond as long as the overall objective is [achieve x].'

Still, if you know that Jupiter was a fund you 'nearly bought' then it may do you some good (for better or worse) to continue to monitor how you would  have done if you had used it.

Most professionally researched models would probably not use '100% strategic bonds' as their non equity allocation, but may allow some portion to float as 'strategic' while making deliberate allocations to other classes (e.g. high yield, gilts, IL/TIPs etc) based on the modelling they have done. 

aroominyork

bowlhead99 said:

aroominyork said:

There is no one-size-fits-all solution but it seems good enough to use Jupiter Strategic Bond, which I consider a well-managed and relatively conservative fund, as my point of reference.

I generally don't think that any one fund in a sea of rival funds is a great benchmark because they will all go about their allocations in a different way, having different views on the market and tactics to employ to achieve the objective. So one year, well done you might have beaten Jupiter but perhaps Jupiter was in the bottom quartile. A strategic bond sector average would be more useful if you are not going to take a view  for yourself yourself on how to position your holdings and would in practice let a bond manager allocate all the money you give him with free rein to pick 'any kind of bond as long as the overall objective is [achieve x].'

Still, if you know that Jupiter was a fund you 'nearly bought' then it may do you some good (for better or worse) to continue to monitor how you would  have done if you had used it.

Most professionally researched models would probably not use '100% strategic bonds' as their non equity allocation, but may allow some portion to float as 'strategic' while making deliberate allocations to other classes (e.g. high yield, gilts, IL/TIPs etc) based on the modelling they have done. 

Yup, I get that. Better may be to take the performance of the Sterling Strategic Bond sector off a Trustnet chart.

2unlimited91

Benchmarking against an active fund (or against a sector of mostly active funds) seem a bit odd.

I think using a global aggregate bond fund, hedged to sterling, as a benchmark, has some merits. It's a relatively simple answer. Most UK-based investors would use sterling or sterling-hedged bonds.

If you don't like that, you could build your benchmark from a small number of passive components. E.g. 2/3 in a gilts index fund and 1/3 in a corporate bond index fund (where the latter could be either a global corporate bond index fund hedged to sterling, or alternatively a sterling corporate bond index fund). I picked 2/3 and 1/3 because that's roughly what global aggregate bond funds give you for sovereign vs corporate bonds, but you could change the proportions to taste.

As a variation, you might want to include cash as a third component. After all, that is something that strategic bond funds may also be holding. (Then you have to pick some way to measure a "cash index".)

aroominyork

I just charted the Sterling Strategic Bond sector (B, yellow) against the Global Bond sector (A, green) and Vanguard's hedged Global Bond Index (C, red). Is the difference between the latter two down to C being hedged and A unhedged, or are there other significant factors at play?

Sailtheworld

I don't know if this is a stupid question but if you went passive why would you need a benchmark - you'd be invested in it and be guaranteed to achieve it minus costs.

aroominyork

Sailtheworld said:

I don't know if this is a stupid question but if you went passive why would you need a benchmark - you'd be invested in it and be guaranteed to achieve it minus costs.

I'm not. I'm looking for a benchmark for my mixed active/passive portfolio. If I went fully passive I wouldn't need a benchmark.

aroominyork

Decision made. For the equity component of the benchmark I am using both HSBC FTSE All World Index and VLS100 with cells to adjust the proportions, eg 50/50, 60/40. And for fixed interest I went with the Sterling Strategic Bond sector from Trustnet. I then managed to create the formula for the benchmark: 

=(((1+J79)*I80*$F$43)*(F80/F79))+(((1+J79)*I80*$G$43)*(G80/G79))+((1+J79)*(1-I80)*(((1+H80)/(1+H79)))-1). One day I'll try to simplify that!

bowlhead99

aroominyork said:

I just charted the Sterling Strategic Bond sector (B, yellow) against the Global Bond sector (A, green) and Vanguard's hedged Global Bond Index (C, red). Is the difference between the latter two down to C being hedged and A unhedged, or are there other significant factors at play?

There are two factors at play. One big one is currency as you mention.

As you can see, a little over two-thirds of the way through the 'Jun15 - Dec 16' section (so, late June 2016) the Global Bond sector jumps through the roof, and by mid November 2016 you can see that about £140 in global bond funds held in the middle of June would generally be worth £160. That's broadly consistent with the fact that pre-referendum, $200 was worth £140, and towards the end of that year $200 was worth £160, due to a change in exchange rates. Most global bonds are denominated in dollars ; some will be in other currencies which did not appreciate  quite as much against sterling, such as Euro, but we generally know sterling devalued and global assets became worth more sterling.

The Vanguard fund is hedging out its currency exposure so that in June 2016 when sterling devalues, and the the chart for global bonds converted to GBP goes vertical, the hedged index simply has a mild bump - in line with what happened to the underlying bond values in their own currencies. I haven't looked at how accurate the hedging was, though hedging is never 100% perfect if being done at acceptable cost. But currency effects will hinder the relative return of the hedged Vanguard fund when sterling weakens and help it versus alternative choices when sterling strengthens.

The other factor is that the Investment Association Global Bonds sector line is simply the average return of the funds being offered in the UK market that meet the criteria for inclusion in that fund sector rather than some other fund sector. Those UK managers are only a subset of the people who are allocating $trillions to listed bond investments across the globe - IA Global Bonds sector was measuring £43bn of funds at March 2020 - and on average may not choose to allocate their investments across all the opportunities that exist weighted to value.  So, you wouldn't expect the IA average to give the same as the Vanguard fund even if the Vanguard fund was unhedged. The Vanguard fund is not trying to deliver the average of the other UK IA Global Bonds sector funds. 

Really your benchmarking decision (which it sounds like you have resolved now, from your latest post) goes back to making a decision on whether you want to benchmark your returns to :
- some weighted average of what bonds happen to exist (Bloomberg Barclays Global Aggregate Float Adjusted and Scaled Index in GBP)
(hedged to sterling, or not hedged to sterling); or

- what global bond funds happen to be offered (IA Global Bond sector average); or

- what general fixed income funds happen to be offered (IA total Fixed Income, being a sum of their sterling Corporate Bond, sterling high yield, sterling strategic bond, global bond, global emerging markets bond, UK gilts, UK index linked gilts sectors; this covers £210bn of bond fund products vs £43bn in just the Global Bond sector); or

- what particular types of bonds you would actually be interested in buying (a weighted mix of different classes of bonds you would like to hold, blending the IA sector returns together accordingly); or

- what specific 'rival' bond funds you would have bought if you had not bought the ones you did.

By all means, use a Vanguard global bond product (hedged) or an iShares global bond product (not hedged), or an IA global bond sector average, as your comparative. But you already know it would be tracking a mix of holdings you don't want. 

If the only type of bond fund holding you would entertain is a 'strategic bond' fund where someone else decides how to invest your money because you can't decide how much to invest in corporate, government, high yield, EM, index linked etc, then there is some logic behind using the sterling strategic bond sector average as your benchmark. But that is tracking 'how are other people getting on when they give their money to strategic bond fund managers this year', rather than 'how much return is available from bonds this year'.