≡ Menu

Which commodities ETF?

For the final installment of Monevator’s commodities series, I’m going to walk you through my ETF picks for actually investing in this controversial but potentially highly useful asset class.

Of course the recent history of commodities investing has been more troubled than that of an 11-year old arsonist.

We’ve taken four posts just to lay out the pros and cons:

While reasonable minds might disagree on this one, my own conclusion is investing in broad commodities offers me portfolio diversification advantages I can’t get elsewhere. 

And now that I’ve researched the available commodities ETFs, I’m satisfied I should be able to pick up a product that can capture the benefits of the asset class, too.

I’ll dive into those details further down.

But first let’s run through my candidates for the best commodities ETFs.

The best commodities ETFs that I can find

My top pick is the nattily named: 

UBS ETF (IE) CMCI Composite SF UCITS ETF (USD) A-acc

  • Ticker: UC15
  • ISIN: IE00B53H0131
  • Index: UBS Constant Maturity Commodity Index
  • OCF: 0.34%
  • Launched: 20 December 2010
  • Domicile: Ireland
  • Replication: Synthetic 
  • Cumulative nominal GBP return: 18.3% (20 December 2010 to 6 July 2023)

This is the ETF I’m going to invest in. 

My alternative choice is:

L&G Longer Dated All Commodities UCITS ETF

  • Ticker: CMFP
  • ISIN: IE00B4WPHX27
  • Index: Bloomberg Commodity 3 Month Forward Index
  • OCF: 0.3%
  • Launched: 18 March 2010
  • Domicile: Ireland
  • Replication: Synthetic 
  • Cumulative nominal GBP return: 9.4% (20 December 2010 to 6 July 2023)

Both these broad commodities ETFs track an index that’s locked on to the fortunes of energy, agriculture, livestock, and industrial and precious metals futures contracts. 

That’s a mouthful but it’s what we want to see. (We covered the particulars in a previous commodities investing post.) 

Essentially, we want exposure to a diversified range of commodities futures. That’s exactly what we get with these two ETFs. 

Here’s the battle of the commodities index trackers across the maximum comparable timeframe:

A chart showing our top two commodities ETF picks head-to-head

Source: justETF.

UC15 (red line) comes in comfortably ahead of CMFP (blue line) with a cumulative return of 18.3% compared to 9.4%.

Though do note the -45% cumulative return chalked up by 2016!

UC15’s better overall return makes it my top pick. But as you’ll see shortly there are reasons you could choose CMFP instead.

There’s certainly no guarantee that UC15 will continue to beat CMFP in the future.

Why did I choose these commodities ETFs?

My main criteria for my best commodities ETF choices are that they:

  • Track an index that can capture the benefits of the asset class
  • Have a good long-term track record (relative to mainstream commodities benchmarks)
  • Have a reasonably low cost
  • Don’t do anything weird (relative to other broad commodities trackers)

My top two picks tick all those boxes. 

I started with the universe of London Stock Exchange broad commodity ETFs listed on justETF, excluding currency-hedged ETFs. As with equities, I want the currency risk associated with commodities (which are denominated in US dollars). 

My next stop was to establish which ETFs are doing a decent job of capturing the benefits of the broad commodities market. 

My benchmark here is the Bloomberg Commodity Index (BCOM). This is the contemporary version of the investable index I’ve used throughout this series to establish that broad commodities are a worthwhile long-run investment.  

(The other well known commodities index is the S&P GSCI, but not a single European ETF tracks it.)

Now, commodities have endured a terrible bear market for most of the period these ETFs have been around. So prepare for your drooling chops to dry when I start bandying around numbers. 

Remember we’re here because future expected returns for commodities are estimated to be 3.5% to 4% (annualised real total returns). And because the historical real GBP returns are 4.5% annualised over the past 89 years – trouncing any other asset class that can diversify our equity returns.  

Long-term track record

We want our commodities ETF of choice to have matched or beaten the BCOM total return index over the longest possible timeframe. 

Both of my top picks launched in 2010, so that’s plenty of time to ascertain whether they work or not. 

For comparison, the BCOM total return index delivered a stonking -2.8% nominal cumulative total return (GBP) from 2011 to 2022. (Take me home, mama!)

  • UC15 earned a 26.7% cumulative return over that period 
  • CMFP earned 21.08%

The point is both ETFs smashed the BCOM return over 12 years. 

Which brings me to the next key point…

Your commodities index matters

When you pick a global tracker fund, the index matters, but not that much – just so long as it’s a reputable global equities benchmark. 

However we can’t be so complacent with our broad commodities index. 

Broad commodity indexes are divided into first generation, second generation, and – oh happy day – third generation iterations. 

Second- and third-gen indexes fix some of the known problems with first generation indexes.

Yet the two most popular benchmarks (BCOM and S&P GSCI) are both first-gen originals. 

Antti Ilmanen sums up why this commodity index innovation has been a positive in his book Expected Returns on Major Asset Classes:

Nonetheless, the changes made in the new indices appear a priori reasonable: less weight in the energy sector; changes in roll schedules (because monthly rolling from nearby to second contract according to the S&P GSCI’s schedule puts one-sided pressure on market prices and causes temporary price distortions that other traders can exploit or avoid); and, increasingly, a shift from holding only the most liquid nearby futures contract toward including a basket of deferred contracts. 

That last point is particularly important.

Because first-gen indexes only track the most liquid short-dated contracts, their returns typically suffer in contangoed markets. 

Some second-gen indexes mitigate the problem by including contracts further along the curve. 

The indexes followed by both UC15 and CMFP use this technique. And these ‘curve management’ strategies actually work, according to Adam Dunsby and Kurt Nelson in A Brief History Of Commodities Indexes:

By distributing positions across the curve, investors have mitigated this impact and achieved higher returns.

Clearly this approach has paid off for UC15 and CMFP as they’ve both easily outperformed the first-gen BCOM index since launch. 

You’ve been contangoed

Sadly, we can’t assume the second-gen indexes will always win. As Dunsby and Nelson explain, UC15’s second-gen UBS CMCI index is better in certain conditions:

When contango is more pronounced in the front end of the futures curve, as is typically the case for, say, corn and has recently been the case for crude oil, then these indexes will outperform the first-generation indexes. When futures markets are backwardated, and the backwardation is concentrated in the front end of the curve, then these indexes will underperform the first-generation indexes.

As for CMFP’s BCOM 3 Month Forward index, Dunsby and Nelson say:

The DJ-UBSCI [BCOM] 3 Month Forward takes a different approach. It invests in the commodities contracts that the traditional DJ-UBSCI would hold three months from now.

This feature places all the DJ-UBSCI F3 contracts farther out the futures curve, and since futures curves tend to be flatter as tenor is extended, the effects of backwardation and contango tend to be reduced. 

The market tends to switch between backwardation and contango, but contango has dominated returns over the period since most commodity ETFs were launched. 

The risk in plumping for UC15 is that we end up kicking ourselves as a golden age of backwardation inevitably follows, simply because we tried to outwit fate. 

Neuroticism aside, the more even-handed approach of CMFP could make sense, despite its less impressive overall return, given that predicting the futures curve is well above our paygrade. 

Finally on indices, just in case you’re wondering, Rallis, Miffre, and Fuertes say in Strategic and Tactical Roles of Enhanced-Commodity Indices:

Our findings suggest that the enhanced indices retain the risk diversification and inflation-hedging properties of the traditional S&P-GSCI and DJ-UBSCI [BCOM index]. 

Commodities ETF alternatives

There is a cheaper way than CMFP to track the BCOM 3 Month Forward Index – namely Xtrackers Bloomberg Commodity Swap UCITS ETF 1C, or XCMC for short. 

XCMC’s OCF is only 0.19%, versus 0.3% for CMFP.

The reason I haven’t given XCMC the nod is because it only launched in November 2021. 

That said, the two ETFs have been neck-and-neck over the 20 months XCMC has existed. If you don’t mind a short track record, then choosing XCMC on cost grounds looks reasonable as so far it’s hugging its index as effectively as CMFP.

Elsewhere, Xtrackers Bloomberg Commodity ex-Agriculture & Livestock Swap UCITS ETF 2C (XBCU) actually pipped UC15 in a dead-heat for the period 2011 to 2022.

Personally, I don’t want to exclude agriculture and livestock. But if you do then this is the ETF to consider. It follows the Bloomberg ex-Agriculture and Livestock 15/30 Capped 3 Month Forward index. 

The UBS ETFs (IE) Bloomberg Commodity CMCI SF UCITS ETF (USD) A-acc (UD07) looks promising but it only launched in 2017. This ETF follows a slightly different 2nd-gen constant maturity index from UC15. 

Amundi Bloomberg Equal-weight Commodity ex-Agriculture UCITS ETF Acc (CRBL) is a contender and the only broad commodities ETF available that tracks an equal-weighted index. However, the ETF changed its benchmark in January 2023 making it difficult to know whether its long-term returns are still relevant. 

I passed on the Invesco Commodity Composite UCITS ETF Acc (LGCF) for the same index tinkering reasons. 

Finally, if you want a first-gen index tracker then check out Market Access Rogers International Commodity UCITS ETF (RICI). It’s consistently outpaced BCOM since 2007. 

Of course, you could always divide your commodities’ allocation between a couple of meaningfully different approaches.

You could split your money 50:50 between a first-gen and second-gen ETF.

Or between a second-gen and the equal-weight ex-agriculture option. Academic research into commodities futures shows that equal-weight indexes have historically outperformed their first-gen counterparts.

It depends on your tolerance for portfolio complexity.

Commodities ETF mop-up

ETFs are not covered by the FSCS investor compensation scheme, although you’d still be eligible for support if your broker went bust. 

If you invest outside of your tax shelters then make sure your commodity ETFs have UK reporting fund status. Otherwise capital gains will be taxed at your marginal rate of income tax. 

If an ETF’s assets under management (AUM) are worth less than $100 million a couple of years after launch, then it may eventually be closed down or merged with another fund. 

That doesn’t mean you lose your money, but it can leave you out of the market for a while, or trigger a tax event – potentially annoying if you’re investing outside an ISA or pension. 

Note that my top two ETFs are both comfortably over $100 million in AUM. 

Commodities ETFs don’t pay dividends but they do reinvest interest earned on collateral. Consult a tax professional if this concerns you. 

I haven’t looked at funds that invest in commodity stocks because they are highly correlated with the broader equities market. You need to invest in broad commodities ETFs to get exposure to the diversification benefits we’ve examined in this series. 

Swap shop

Commodities ETFs use total return swaps to track their indexes. A total return swap is a derivative, provided by a third-party who undertakes to pay the ETF the return of the index (minus costs).

This arrangement means commodities ETFs don’t actually invest in futures contracts, never mind shipments full of lean hogs, bales of cotton, or barrels of oil. 

Index trackers that use total return swaps are classified as synthetic ETFs and it’s as well to know what that entails. 

In reality, each synthetic ETF’s holdings amount to a basket of securities that have nothing to do with commodities and are held as collateral. This is standard practice.

The collateral is there to cover investors against counterparty risk – the chance that the swap provider fails to pay out during some kind of financial crisis. 

On that basis, make sure to check that your chosen ETF’s website indicates the tracker is backed by collateral worth at least 100% of its market value.

Commodities ‘coaster

Broad commodities had an awesome 2021 and 2022. But the asset class is hovering around correction territory (-10%) so far in 2023. 

If that gives you the heebie-jeebies then commodities are not for you.

They suffer equity-scale volatility and can spend years – even decades – underwater. 

Eruptions into positive territory can be brief but spectacular, like watching the geyser Old Faithful blow its spout in Yellowstone Park. 

But there’s no point investing in commodities if years of negative returns turn you into Old Unfaithful – ditching your holding and so failing to collect when returns sky rocket. 

Personally, I’m still nervous about commodities. But I’ve been persuaded by the long-run data that shows the asset class can work when others fail. 

For this reason, I’m going to take the plunge now I’ve homed in on a couple of commodities ETFs that look up to the job. 

That said, I will proceed cautiously. I’ll start by switching a few percentage points of asset allocation, then build up my position slowly over the next couple of years, or whenever the market takes a downward lurch.

With commodities – more than my other asset classes – I want to minimise any early regrets, get comfortable, and then hunker down for the long-run.

Take it steady,

The Accumulator

More ETfs for fun and profit:

{ 68 comments… add one }
  • 1 MCM July 25, 2023, 11:33 am

    Why would you even buy this kind of volatility?

  • 2 JohnP July 25, 2023, 12:28 pm

    Thanks for a great set of articles, regarding asset allocation, I currently have a basket of funds, that roughly break down to.

    Global Equity 76.82%
    Property 2.56%
    Gold 4%
    Mixture of bond funds 16.62%

    If you was going to allocate some of this to a broad commodities fund, do you have any suggestions? I was thinking about dropping property and or reducing equity a bit to fit commodities in. As I already have a selection of REITs in a more actively focused portfolio

  • 3 The Accumulator July 25, 2023, 2:58 pm

    @ MCM – Because commodities are negatively correlated with equities and bonds that volatility makes it a great diversifier:

    https://monevator.com/why-commodities-belong-in-your-portfolio/
    https://monevator.com/commodities-diversification/

    Of course, it’s only bearable if you can view the portfolio holistically and the asset class delivers something like its historical long-term returns.

    @ JohnP – Without knowing anything about your objectives or risk tolerance etc it certainly doesn’t look like you’re overburdened with bonds. If you want to maintain the same equities allocation then you could decide to swap the gold in for commodities given they overlap strategically.

    Assuming your property holding is an index tracker full of global REITs then it’ll be highly correlated to your global equities and thus seemingly dispensable. Although TI has rightly argued in the past that REITs still provide some diversification value versus equities.

    Otherwise, the equity allocation does look like the place to carve out a chunk. Sounds like that’s your intuition?

  • 4 Time like infinity July 25, 2023, 3:02 pm

    Interesting. Thank you @TA for this superb and intriguing quintet of articles. There’s much here to think through.

    Obviously, 89 years’ of data is better than any subset within the 1934-2022 span, but in terms of ‘expectation management’ for future returns would it perhaps make sense to also look at the returns for, say, the 1950-2016 period?

    My thinking here is that way back in 1934 commodities presumably we’re coming off a massive low with the nadir of the Depression, and by 2022 they’d already rallied hard, up circa three fold (just eyeballing the chart) since 2016. I’m guessing that narrowing the date range could possibly knock a couple of percentage points annually from the 4.5% p.a. real GBP return when comparing the 1950-2016 timeframe to the whole 1934-2022 period.

  • 5 JohnP July 25, 2023, 3:27 pm

    @TA thanks for your thoughts on the matter.

  • 6 Hobard July 25, 2023, 3:59 pm

    What % allocation would you recommend for a 90-10 equity vs bond portfolio? I appreciate the answer is “it depends”, but I am curious to understand how much people are generally thinking of adding this as an equity diversifier

  • 7 Laurence July 25, 2023, 4:42 pm

    Just wanted to say as a long time reader, first time commenter that I have really enjoyed this series and the outcome is intriguing.

    I’m always looking for diversifiers, outside bonds, particularly during the ZIRP years. I have dabbled with long-short funds, infrastructure funds, private equity, Aircraft leasing, trend following and arbitrage strategies to name a few! The main issue / opportunity with lots of these investments is they are often Investment Trusts which seem to follow the wider market, rather than their underlying assets in times of stress.

    I historically stayed away from commodities due to their high volatility (and believed my allocation would be covered by mining stocks etc.) , and admittedly mainly looked at gold, which seemed too correlated to interest rates for my liking. I’m thinking of putting 10% of my portfolio in commodities and continuing to research the topic.

  • 8 teamdave July 25, 2023, 5:26 pm
  • 9 Meany July 25, 2023, 6:04 pm

    re
    >BCOM total return index delivered -2.8% from 2011 to 2022.
    >UC15 earned 26.7%

    Is the BCOM index used here also unhedged and held in £?
    If not, the gain on UC15 is just the $1.50->$1.29 currency drop.
    (If so, the -2.8% on commods is about -25% hedged!)

  • 10 The Accumulator July 25, 2023, 6:54 pm

    @ Meany – All returns quoted are GBP. Hedged isn’t the way to go IMO.

    @ Teamdave – BOCG tracks BCOM. I think the simplicity of the first-gen indexes generally means lower fees but I also think picking the right index is more important with commodities than with equity trackers.

    @ Laurence – cheers! And great that you’ve broken your commenting duck 🙂 Completely agree about the search for a diversifier that works when bonds don’t, though I haven’t played the field as much as you!

    @ Hobard – check out this piece for a useful summary of the research into asset allocation: https://monevator.com/commodities-diversification/

    Personally I’m going for 10%. Less won’t move the needle but I’m just not brave enough to go as high as 20%.

    @ TLI – Thank you, and a very thoughtful provoking comment as always. I certainly think it’s a good idea to rein in expectations versus historical average returns for all asset classes. The consensus on future expected real returns is around 3.5% to 4%.

    The 1930s don’t look that bad with the gain in the 1940s being driven by WW2-related inflation. You’re right though, knock 2% off if you prefer conditioning on that time period. Certainly I wish I’d uncovered all this around 2019-20.

  • 11 Time like infinity July 25, 2023, 10:00 pm

    Pondering SRI/ESG implications of UC15 having 18.2% in West Texas Intermediate crude oil (the NYMEX & ICE contracts are its top two holdings, at 9.5% and 8.7%). Previously, I’ve taken an inclusive ‘whole of market’ approach to trackers, and not ‘boycotted’ sectors, basically on pragmatic grounds, e.g. does it actually make much difference not to own securities (via a tracker) traded on the secondary market when it’s by share issues that companies raise capital; and how exactly to define SRI. However, it feels a bit uncomfortable to be thinking about buying an ETP which is so heavily into fossil fuels. Perhaps it’s just recent coverage, but we do seem at a turning point:
    https://www.theguardian.com/commentisfree/2023/jul/25/frightens-climate-crisis-do-not-know-how-bad-wildfires-greece

    I’ve no problem at all with any industrial metals. For one thing, we’ll need plenty for the green transition:
    https://www.thedriftmag.com/a-good-prospect/

    But the concentration in oil is troubling.

  • 12 Sparschwein July 25, 2023, 10:54 pm

    Thanks for the interesting series. This made me re-consider futures ETFs instead of the producer stocks.

    Have you looked into ETFs with “optimised roll” and how they compare with second-gen indexes? Here in a comparison from 2019 the optimised roll seemed to do better than the Bloomberg 3 Month Forward Index (see figure 4; caveat – marketing material)
    https://www.wisdomtree.eu/-/media/eu-media-files/other-documents/research/strategy-insights/wisdomtree_strategy-insight_how-to-optimize-a-commodity-portfolio_en.pdf

    There’s an Enhanced Roll Yield Index from Bloomberg too, with available ETFs e.g. from iShares.

  • 13 Paul_a38 July 26, 2023, 7:08 am

    Good article. I put some into the l&g ETF, not enough probably, some time ago based on articles elsewhere. L&G offer an enhanced commodity fund ( ENCG) but as I understand even less about this than the conventional approach I balked.

  • 14 The Accumulator July 26, 2023, 9:53 am

    @ Sparschwein – Yes, I checked out the WisdomTree Enhanced Commodity ETF and iShares Bloomberg Enhanced Roll Yield Commodity Swap ETF.

    They both significantly lag UC15 and CMFP over the maximum comparable timeframe using live data.

    WisdomTree use backtested data in their material. Quelle surprise, they managed to choose dates that favoured their product. The Investor posted a link to a good Morningstar piece on how funds often fail to live up to their backtested billing:

    https://www.morningstar.com/etfs/some-index-funds-are-riskier-than-you-think

    The iShares ETF lagged and also only switched to that particular Bloomberg index in Feb 2022. I’d guess because its previous Bloomberg optimised roll index wasn’t delivering the goods.

    Optimised roll is a perfectly legitimate second-gen index approach according to the independent research I’ve read. All we can conclude, I think, is that the live data shows optimised roll hasn’t conferred any advantage so far on the ETFs that use it.

    @ TLI – Remember all these ETFs are synthetic. They’re just buying in the index performance using a total return swap. They don’t physically hold the contracts, but they do show you the composition of the index. From an ESG perspective, it’d be worth checking their collateral holdings – which are physical – to see if there are any red flags there.

  • 15 Corkscrew July 26, 2023, 5:32 pm

    I’ve enjoyed this series. I’m still a little wary of commodities because of the complexity/ nuance involved and also the fact that funds haven’t been available for huge periods of time.

    Any thoughts on the fact that correlations between commodities and equities might have increased since 2008?(not being a good thing and making them less appealing) Haven’t done a super deep dive into the data, but looking at US equities and PCRIX seems to suggest recent correlations have been less appealing

  • 16 The Accumulator July 27, 2023, 10:29 am

    @ Corkscrew – cheers, and I share your wariness. The post 2008 correlation is 0.36 which is roughly what the long-term correlation between gilts and equities is – so still a good diversifier.

  • 17 Corkscrew July 27, 2023, 12:30 pm

    @TheAccumulator – Thanks for the info. What do you use to calculate the correlation across time periods? I’ve been playing around with https://www.portfoliovisualizer.com/asset-correlations but it only has US funds and would like to play with the numbers for UK equivalents

  • 18 Eagleuk July 27, 2023, 3:23 pm

    UC15 is not listed at trading212.They have etc ticker CMCU available for trading.

  • 19 JohnP July 27, 2023, 3:36 pm

    @Eagleuk if you have not already tried I would try asking them to add it, they have a stocks requests section on their forums.

  • 20 The Accumulator July 27, 2023, 4:05 pm

    @ Corkscrew – You can download BCOM Total Return index from here:

    https://www.bloomberg.com/professional/product/indices/bloomberg-commodity-indices-fact-sheets-and-publications/

    BCOM Historical Index Levels > BCOM, Sectors, Single Commodity Indices

    I originally converted USD returns into GBP but have since found GBP numbers using the above link.

    To go back further in time use the data appendix from The First Commodity Futures Index of 1933 paper. You will need to convert from USD returns. USD/GBP exchange rate from BOE / Millennium of Macroeconomic Data.

    You can get UK equity annual returns from Macrohistory Database and then FTSE Russell for the last couple of years.

    I’ve used annual return correlations.

  • 21 David C July 27, 2023, 6:26 pm

    Thanks for all your work on this. As far as I can see Halifax/BoS don’t carry the UBS UC15 ETF, which presumably means Lloyds and iWeb don’t either. They do have the L&G CMFP ETF. Fidelity FundsNetwork don’t have either. That’s assuming they haven’t listed them under a mangled name; it wouldn’t be the first time.

  • 22 Daniel July 27, 2023, 6:55 pm

    Great article series, have been tempted by commodities for a few years but felt too uneducated to invest.

    I did put a little position in a few months ago and choose iShares Diversified Commodity Swap UCITS ETF.

    Any thoughts on this etf and how it compares with your choice?

    https://www.justetf.com/uk/etf-profile.html?isin=IE00BDFL4P12&from=search#overview

  • 23 Mr Optimistic July 27, 2023, 7:19 pm

    @David C. Halifax et al are pretty good at adding a listing if you request it. They added something for me in a couple of days.

  • 24 The Accumulator July 27, 2023, 8:54 pm

    @ David C – I couldn’t find UC15 at Fidelity either but AJ Bell have it.

    @ Daniel – yes, I looked at COMM. It tracks the 1st gen BCOM index pretty closely. Any ETF that tracks BCOM would have lagged the choices I’ve highlighted by some way. That may not continue in the years ahead depending on the prevailing state of the futures curve – which I don’t think we can hope to predict. Still, the evidence so far is that the 2nd gen index trackers are doing a better job than 1st gen ETFs. Among the 1st gen trackers, I found that RICI did well despite its high cost.

  • 25 Serg July 28, 2023, 3:35 pm

    Why so expensive ETFs and not this one with only 0.15 % charge?
    https://www.justetf.com/uk/etf-profile.html?isin=IE00BF0BCP69&from=search#overview

  • 26 The Accumulator July 28, 2023, 5:25 pm

    @ Serg – BCOG tracks the BCOM index which has lagged my 2nd-gen index choices. Since its inception, BCOG has cumulatively returned 41% vs CMFP @ 69% and UC15 @ 77% over the same timeframe. The higher costs for 2nd-gen index trackers appear to be worth it. Though see the index section of the article and my answer to Daniel in #24 above for caveats.

  • 27 Sparschwein July 30, 2023, 12:00 am

    @TA – thanks for the information. There seems to be a good case that UC15 improves the roll yield problem (one of the main reasons I opted for commodities stocks instead). Why UC15’s methodology works and the optimised roll ETFs do not is beyond me though, and this lack of understanding the mechanics makes me uneasy.

    I’ll probably split 50/50 between UC15 and another 2nd or 3rd gen index.
    From the historical data I’ve seen, backwardation is the exception and contango is the “natural” state, which makes fundamental sense because of storage and shipping costs. So FWIW I see no point going back to a 1st gen index hoping for the golden age of backwardation.
    I’d also consider an active funds in this case, if one with a good track record is available. There is FTGC, but US domiciled.

  • 28 The Accumulator July 30, 2023, 1:11 pm

    @ Sparschwein – Completely agree about 1st-gen indices. Re: 2nd-gen, I can’t find anything in the literature to suggest that optimised roll indices aren’t viable. It could just be that market conditions have favoured the constant maturity approach during the relatively short live period these things have been live.

    Or it could be that some variation in index commodity weightings has proved the difference.

    The closest apple for apples comparison we can do for constant maturity vs optimised roll takes UC15 out of the equation as it uses UBS’ CMCI methodology.

    Instead we can pit UD07 (BCOM Constant Maturity Commodity Index) vs ROLG which has used two different BCOM optimised roll indexes.

    Max timeframe goes back to Oct 2018. Constant Maturity index wins by 8% cumulative. If constant maturity trackers maintain that edge over the long-term then that’s the place to be, but perhaps they won’t maintain the advantage – your 50:50 split idea seems sensible given the uncertainty.

    I guess we could use the richer US range of products to create more optimised roll vs constant maturity match-ups but, well, perhaps that’s a thrill-ride I’ll save for another day 😉

  • 29 Sparschwein July 30, 2023, 6:56 pm

    What could be more exciting on a rainy Sunday afternoon…

    From Bloomberg’s index factsheets:
    ROLG’s previous index (BCOMRST) vs. the original BCOM:
    0% vs -1% p.a. over the past 10 years to end June 2023.

    By a separate calculation from yearly performance data 2013 – 2022 taken from index factsheets, all percentages not annualised:
    BCOMRST: -2%; the difference to the above is probably from the shifted timeframe
    BERY or BERYTR (unclear from which the data was) = ROLG’s new index: +14%
    UBS CMCI index = UC15 ETF: +16%
    UBS BCOM Constant Maturity Index = UD07 ETF: +6%

    So a 50/50 mix of UC15 and ROLG it is, case closed.

  • 30 Ben July 31, 2023, 12:23 pm
  • 31 Joseph August 2, 2023, 3:04 pm

    Hello!

    I wanted to ask, what do you think of these two ETFs?

    L&G Multi-Strategy Enhanced Commodities UCITS ETF – IE00BFXR6159
    Invesco Bloomberg Commodity UCITS ETF Acc – IE00BD6FTQ80

    Would the Invesco be a good 1st-gen option, like the COMM and RICI?

    Thanks for the series, quite insightful!

  • 32 The Accumulator August 3, 2023, 8:25 am

    Hello Joseph!

    The Invesco ETF (CMOP) is behaving much the same as other BCOM trackers such as COMM and BCOG.

    They’re all tracking the index pretty well but BCOG is marginally better – probably due to cheaper fees.

    RICI has outperformed by some distance though. A cumulative return of 49.5% versus 27% for the BCOM total return index itself (2007-2022).

    The L&G Multi-Strategy ETF (ENCG) looks interesting but it’s very new so we don’t have much evidence to go on. It has performed well in the two years since launch though: 53% cumulative vs 40% for UC15.

    Two years data isn’t enough for me but I’ll be keeping an eye on it.

  • 33 Algernond August 3, 2023, 12:01 pm

    My ETF momentum strategy got me to buy WisdomTree Energy Commodity ETF (AIGE) this week.

    One thing that I can’t quite get my head around is how much ‘Assets Under Management’ matter for commodity ETFs. For the physically replicated stock ETFs, I always try to buy the ones with higher AUM (e.g. > $500M) to make sure adequate liquidity / lower spread in times of need. But not sure how much this matters for a swap based ETF..?

  • 34 Time like infinity August 3, 2023, 1:18 pm

    Interesting @Algernond. I wonder if this signals higher wholesale energy prices in expectation of winter. Started to think about whether one could do a variant of dual momentum with a permanent 60% allocation to a global equities ETF but with the balance of 40% then put into either a broad commodities ETF or into a AAA-AA rated Global Government Bonds hedged to GBP ETF according to whichever of commodities or bonds had the highest relative momentum on a 6 month look back period and rebalanced every 6 months. That means that all the benefits and risks of being permanently invested in diversified equities are kept but that there is:
    1. a choice of asset classes between commodities and bonds to try and diversify away some of the equity risk; and
    2. instead of trying to guess in advance which of commodities or bonds would better fulfill that function one instead lets momentum make the decision.

  • 35 Time like infinity August 3, 2023, 4:57 pm

    Just ran the numbers on idea @ #34: with 0, 1 or 2 possible switches p.a. Using UC15 or IGLH (the cheapest hedged Global Gov Bond ETF on HL), and with 6 monthly review periods, the round trip costs using quoted indicative spreads are 0%, 0.37% or 0.74% p.a., so an average of 0.37% p.a. (comprised 0.25% one way on UC15 and 0.12% one way on IGLH). To this obviously has to be added the OCF/TCO for the ETFs involved themselves.

  • 36 Algernond August 3, 2023, 5:06 pm

    @Time like infinity – I may go to an easier to manage approach such as the one you outline above at some point in the future. For now with Commodity ETFs, I’m happy to buy them when my Algo tells me to down to various sector levels (e.g. Energy, Agri, Industrial metals).
    At the moment, my rules don’t allow me to by single commodity ETFs – although I am regretting this with COCO (check it out from last November onwards!!).
    One of the reasons I decided to exclude the single commodity ETFs is because of the relatively small AUMs (e.g. COCO only has ~$9M), hence my post above about whether this matters for Swap based ETFs or not….

  • 37 Sparschwein August 3, 2023, 11:27 pm

    I looked into the L&G Multi-Strategy ETF (ENCG) and got as far as this:
    “Tracks the commodity futures contract whose expiry date is dynamically determined according to a factor specific to the commodity, and tilts its weight…
    …expiry dates of their futures contracts are different from the Bloomberg Commodity Index, and are determined and rolled dynamically based on one of three enhancement strategies, i.e. momentum, seasonality and roll-yield.”

    It has done well vs BCOM and BCOM 3 Months Forward in the back test esp. in the last 3 years. The intransparent and apparently very complex rules make me wonder though, have they just data-mined the back test until they found some random set of rules that happens to look good in the back test.

  • 38 Time like infinity August 4, 2023, 12:44 am
  • 39 The Accumulator August 4, 2023, 11:25 am

    @ Sparschwein – I think back test results have to be treated with extreme caution.

    @ TLS – I’m aware that there is some evidence for risk factors in commodity returns but I hadn’t read that paper. Thank you for sharing!

    Interesting that their excess return number is significantly lower than that found in Commodities for the Long Run yet they use a similar dataset.

    Excess return found in the paper The Commodity Futures Risk Premium: 1871–2018 is similar to Commodities for the Long Run, using a more comprehensive dataset than either Two Centuries or the Long Run.

    Quick chart here: https://summerhavenindex.com/dashboard-beta/

    Dimson, Marsh and Staunton also cite the Long Run and Commodity Futures Risk papers, and use their data in their recent work on commodities (Credit Suisse Global Investment Returns Yearbook 2023).

  • 40 Time like infinity August 4, 2023, 12:32 pm

    Thanks @TA #39 Assume “TLS” = TLI 🙂 There’s timelike, spacelike & lightlike infinities on Penrose diagrams, but timelike space (TLS) is Minkowski 4-D space.

    Agree with you, as DMS’ use of the data indicates, that the CFR & CLR studies are more reliable than Samonov’s/Geczy’s Two Centuries paper.

    I hesitated to post anything to do with Commodities and Momentum as you’d listed Factor Investing as your #1 mistake in your 29 Sep 2021 piece ‘Three Investing mistakes that cost me’ (FWIW I think #3 on your list there is the real biggie, e.g. I absolutely loathe crypto on every level – ideological, technical, financial – but of course I wish, when I’d 1st heard of BTC in late 2011 at $3-$5, that I’d put in 0.01% of my capital to see it do over 10,000x by 2021. If it went to 0, which back in 2011 I mistakenly was almost 100% certain it would, then I’d be down an unnoticeable 0.01%, but if it did what it has, then on 0.01% commited in 2011 it would have added a hefty wodge to whole portfolio by 2021, and indeed still now.)

  • 41 Boffinboy August 4, 2023, 2:16 pm

    Thinking about adding commodities to my ISA portfolio after reading these articles. Currently 70:30 equity:bonds. Anything from 15-25 years before I anticipate potentially being in a decumulation phase, dependent on how my earning potential increases relative to expectations. Was thinking about 5%, but not sure that’s even worth it.

    Happy to keep the chunky equity stake, but slightly concerned about reducing bond cushion by 1/3 if I went up to 10% commodities, especially given some of the long periods of tough performance, and the current upward spike. I suppose I could take 5 off equities and 5 off bonds, or do 7.5%, but still feel slightly wary about it.

    Of course, I’d just simplified my portfolio not long ago to 2 or 3 funds and now I’m tempted to fiddle again. I’m not sure what this says about my psychology… aiming to maximise but in the end probably making things worse by changing.

  • 42 Joseph August 4, 2023, 3:52 pm

    @TA Thanks!

    I see the appeal of RICI when it comes to its outperformance even though it has a higher TER. But what do you think about the low amount of assets under management it has? Same as for BCOG or XCMC, do you think this is something that should be a red flag?

  • 43 The Accumulator August 4, 2023, 6:30 pm

    TLI! Not TLS. Sorry! I was clearly lost in space myself at the time I wrote that. The mnemonic Penrose not Minkowski should keep me on track in the future.

    Re: Crypto, I feel your pain. Except most people I’ve spoken to anchor on whatever paper fortune they’d earned at market peak rather than buy in price. The mind never lets us win!

    I also have a friend who mined crypto in the early days but got his wallet hacked later. He’d have been a millionaire. Can’t get over it.

    @ Boffinboy – I know what you mean about the dissonance of tampering but I think there’s a clear rationale in this case.

    @ Joseph – I don’t think low AUM are a red flag. Just a few marks against that I’d probably apply in a tie-breaker situation. If your fund closes, you’ll get your money back, you’ll just be out of the market for a time. Many ETFs get by on a very small AUM so it’s hard to apply a strict rule of thumb.

  • 44 Joseph August 5, 2023, 7:07 pm

    @TA – I see, though more than the fund closing, I worry about potential illiquidity when in need to sell (for rebalance purposes, for example).

  • 45 The Accumulator August 7, 2023, 8:38 am

    Typically not a problem that besets smaller ETFs to the best of my knowledge – there’s always a market maker out there willing to trade. You might pay a slightly higher spread to trade a smaller ETF but even that doesn’t seem to be universally applicable

  • 46 JohnP August 7, 2023, 1:09 pm

    For anyone that’s interested Trading 212 have now added UC15

  • 47 Chris August 8, 2023, 8:14 pm

    Am I correct in assuming that a 50/50 split between UC15 and CMFP is a sensible approach to diversifying within the commodities assessment class?

  • 48 Joseph August 9, 2023, 12:12 pm

    @ TA – got it, thanks!

    I read this article in the WSJ about the diversification power of commodities and its role in a portfolio. What you think? I guess they are not weighing the 2nd generation indexes into the equation…

    https://www.wsj.com/articles/commodities-investing-boost-returns-a9252c9e

  • 49 The Investor August 9, 2023, 1:04 pm

    @Joseph — I can’t read the article (paywalled) but I imagine it says commodities have done well recently but that over the past few decades they have been very mediocre and also that the connection to inflation is imperfect at best.

    Well guess what, @TA just wrote five articles / 10,000 words saying the same thing 🙂

    People really have to ditch their binary thinking. Investing is complicated.

    Our sometime house troll (he’s currently on best behaviour) accused us the other day of “pushing” commodities because we’d written five articles on the subject.

    No. We published five articles because for exactly the opposite reason — because there’s no yes/no answer and it’s all a bit murky.

    @TA believes recent research shows that over the long term commodities can be good diversifiers and additive to returns. But he’s many times stressed it’s not a slam dunk.

    Investing is always a cloud of probabilities. 🙂

    Even the strong recent returns of commodities can be read both ways – a reason for caution (the good spurt has happened!) or a change of regime and a reason to buy (momentum etc!)

    I’d go very slow and build a small position over several years personally but who knows? 🙂

  • 50 Time like infinity August 10, 2023, 12:02 am

    @TA’s commodity articles are a fantastic 5. A tour d’horizon & tour d’force. There’s a lot to work through & to think about. For those inclined towards pure passive, the articles may throw up a dilemma. 89 yrs’ high quality data shows efficient frontier lay at 60 equities/40 commodo (6% real average annual return, with both a higher Sharpe and annualised returns than for all equities). Some readers, taking Lars’ just 2 assets approach, will be targeting 60 equities/40 intermediate (or long) gilts (or global gov bonds hedged to £). To go from that to no bonds and a whopping 40 in commodo is huge, requiring something of a paradigm shift in thinking. Yet that’s exactly what longest run data indicates they should do in order to optimise for both highest total and risk adjusted returns. Whilst it’s not too uncomfortable shifting from 60 e/40 b to 60 e/20 b/20 c; to go straight to 60 e/40 c is another thing entirely for the cautious. But, whilst commodo allocations less than 40 did reduce risk, they didn’t (over whole of 1934-2022) actually increase the overall portfolio return, as compared to 100 in equities. Market timing buying allocation to broad commodities whilst they are in a deep draw down could, of course, hugely increase the odds that putting full 40 into them will pay off against the alternatives of having that 40 in bonds, or even having the whole 100 in equities. But, as a wise man once said, in investing there are two types of person, those who can’t market time and those who know that they can’t market time. I’m more of a ‘quanty’ semi passive type myself (with a sprinkle of active opportunism); but I see the advantages of the pure passive Bogel way, and it’s clear attractions to others. It seems to me that it could be quite a leap of faith for pure passives to either a). give up their bonds entirely to embrace their inner Markowitz and the new efficient frontier by going for the full 40 in commodo; or b). to try and time their jump into commodo. If they do the latter, then it may be a case of wishing them ‘may the alpha be with you’.

  • 51 The Accumulator August 11, 2023, 11:42 am

    @ Chris – it’d diversify you over the long-term if one ETF happens to best the other by a significant amount, and certainly the indices do take different approaches. Over shorter periods, they’re likely to act similarly – much like my FTSE All-Share fund exhibits the same volatility pattern as my global tracker while its long-term returns over the past 15 years are disappointing in comparison.

    Splitting your allocation between physical gold and broad commodities would be more likely to offer two holdings that can behave quite differently from each other.

  • 52 DaleK September 26, 2023, 4:23 pm

    @TA – Many thanks for the series. After doing my own research I decided to gradually grow a tiny proportion of my ‘defensive’ allocation to commodities (with a target of perhaps only ~2.5% overall eventually).
    Just for info, I was able to buy UC15 from ii last month, but found it was unavailable this month. After enquiring I’ve been told:
    “As part of the FCA’s introduction of Consumer Duty, certain investment companies are now required to undertake Value For Money Assessments (VFMAs) for the instruments they offer. These assessments are focused around ensuring investment companies can justify the charges imposed, and the past performance of the holding, against the relevant benchmark.
    In this instance, the VFMA has not been completed or the investment company has confirmed that the charges cannot be reasonably justified. Therefore, we cannot facilitate purchases in this fund at present. This is an ongoing process and if the assessment is completed in the future, or if the results of the assessment change, then the restrictions may be lifted. ”

    Rather surprising/frustrating! I guess I’ll have to investigate the alternatives…

  • 53 DaleK September 26, 2023, 4:28 pm

    Thanks for the series. After doing my own research I have decided to gradually take on a tiny proportion of my ‘defensive’ allocation over to commodities (perhaps 2.5% overall eventually).
    @TA, just for info I was able to buy UC15 via ii last month, but not now! After enquiring, I’ve been told:
    “As part of the FCA’s introduction of Consumer Duty, certain investment companies are now required to undertake Value For Money Assessments (VFMAs) for the instruments they offer. These assessments are focused around ensuring investment companies can justify the charges imposed, and the past performance of the holding, against the relevant benchmark.
    In this instance, the VFMA has not been completed or the investment company has confirmed that the charges cannot be reasonably justified. Therefore, we cannot facilitate purchases in this fund at present. This is an ongoing process and if the assessment is completed in the future, or if the results of the assessment change, then the restrictions may be lifted.”
    Rather surprising/frustrating! I had better begin researching alternatives…

  • 54 Trevor October 3, 2023, 2:34 pm

    Hello,

    I wanted to ask about the tracking error, because from what I’ve seen, the UC15 lags the index 1% per year. Since inception, the index has returned 17%, while the fund 3%.

    With the CMFP, the tracking error is around 0,4% per year, which is more understandable.

    Do you see this as a problem?

    Thanks

  • 55 The Accumulator October 3, 2023, 5:43 pm

    Hi Trevor,

    It’s definitely not ideal but I can just about tolerate a 1% lag. For me, it’s more about the opportunity set of broad commodities ETFs. I don’t, for example, want an ETF that hugs BCOM like a long-lost brother if BCOM is a flawed index.

    With shares, we can easily find vanilla indexes like the MSCI World or the S&P 500 that are commonly agreed benchmarks for diversified equity exposure. In that instance, it’s a simple case of choosing the tracker that best mimics the index.

    Things aren’t so straightforward with commodities though brand recognition must be a factor as so many ETFs track BCOM.

    Ultimately I’ve name-checked 7 plausible candidates because there’s no standout, problem-free offering.

    I don’t think we can assume UC15’s performance advantage will necessarily persist relative to peers, so choosing CMFP on the basis that it’s a better *index tracker* is perfectly reasonable. As is splitting your commodities allocation between two or three products.

  • 56 Mickey November 10, 2023, 3:47 pm

    Hello @The Accumulator,

    I told my bank I wanted to invest in the L&G Longer Dated All Commodities UCITS ETF, but they told me that it is not possible to invest in it because the fund is closed or is in the process of shutting down. I have not been able to find any information about it. Do you know if the fund is shutting down?

    Thanks

  • 57 The Accumulator November 10, 2023, 4:20 pm

    @ Mickey – I haven’t seen anything about it shutting down. You’d expect there to be a shareholder announcement about the closure but there isn’t:

    https://fundcentres.lgim.com/en/uk/institutional/fund-centre/ETF/Longer-Dated-All-Commodities/

    I’ve tried googling to see if I can turn up a closure announcement in the trade press but nothing there either.

    I’m not saying your bank is wrong but there’s no confirmation as yet. You can see the ETF is still trading daily here:

    https://www.londonstockexchange.com/stock/COMF/legal-and-general-asset-management/company-page

  • 58 Mickey November 10, 2023, 4:46 pm

    Thank you, I will check with them and with the fund, it all seemed odd to me and wanted to check if here you knew something. Hopefully it’s a misunderstanding.

    Cheers!

  • 59 The Accumulator November 11, 2023, 12:51 pm

    Cheers, Mickey. Let us know how you get on. I have an account with AJ Bell and they’ll still let me buy it.

  • 60 Stevey Bee April 25, 2024, 9:14 am

    The KIID for UC15 lists a ‘conversion charge’ of 3%, can anyone tell me what this is?

    Excellent series on commodities, by the way, very educational.

  • 61 GregN August 14, 2024, 2:36 pm

    Thanks, great series of articles. Unfortunately I’m still struggling with the mechanics of how these funds generate positive returns, perhaps specifically through roll return. Is backwardation necessary, or is there a situation where contango also works, and if so, how?

  • 62 The Accumulator August 14, 2024, 4:10 pm

    @ Stevey Bee – Sorry, I missed your comment back in April. I contacted UBS and they say that you won’t be charged this as a retail investor. It’s a compliance technicality apparently.

    @GregN – Contango generally hurts roll returns but positive returns can still be achieved on average. See the table in the ‘When do commodities work’ section here:
    https://monevator.com/commodities-diversification/

    Bear in mind that there are three sources of return for commodity ETFs:
    – Spot price changes
    – Interest on collateral
    – The roll return

    Another thing that’s worth remembering is these types of funds are diversified across commodities futures. Some commodities may be in contango but are offset by others in backwardation.

    Finally, 2nd gen indexes also take steps to mitigate the effects of contango as briefly outlined in this piece.

    Hope that helps. I agree that commodities are not easy to understand!

  • 63 Delta Hedge August 14, 2024, 6:56 pm

    A big thank you @TA for the excellent commodities series which can be read, I think, alongside your recent superb two part All Weather Portfolio pieces.

    Now you’ve got a time series for Global Equity returns going back to 1919 (and possibly to 1900) would you be able to update the equity/commodity mix tables to include Global Equities as well as UK Shares, as currently used?

    I’m really interested to see what a mix of 60 Global Equities / 40 Commodities would show from 1934 in terms of real average CAGR, worst drawdown, average annual volatility and Sharpe Ratio. .

    I’ve added a smidgen of UC15 to my portfolio, and intend to opportunistically and continually build up the position whenever commodities substantially underperform.

    The more information that I have the more confident I am in making strategic asset allocation adjustments.

    Not by design (at least not anymore), I’m radically under diversified in terms of different asset classes (global equities still totally dominate my portfolio); so I need to work much more on commodities, Government bonds, trend following and gold.

    Thank you again for this fantastic five part series

  • 64 Stevey Bee August 14, 2024, 8:10 pm

    @TA Thanks, what actually is the conversion charge though? Is it a currency thing? Or something to do with converting shares to a different type/class? I just couldn’t work it out.

  • 65 The Accumulator August 15, 2024, 11:33 am

    @ Stevey Bee – from the very scant descriptions I could find it looked like it might either relate to financial advisor commissions or as you say some kind of class conversion. But it was very hard to tell. They kept saying refer to the prospectus but that was a wild goose chase.

    @ Delta Hedge – thank you! I would like to do that at some point. I think when this article gets an update that would be a good angle to add.

  • 66 GregN August 15, 2024, 1:34 pm

    Thanks for the clarifications @TA. It’s helped my understanding to see that backwardation in necessary during the relevant timeframe for positive spot return of a particular future. Likewise for the roll return of that future if it is reincluded during the roll period, though it may not be due to rebalancing or reconstitution of the index. This is a feature of the funds/indexes being long only, rather than having short positions. Also worth noting that a fund’s value is continually updated by the spot prices of each of the included futures to reflect these returns, as well as the value of loans held as collateral.

  • 67 Stevey Bee August 15, 2024, 5:59 pm

    @TA Ha, that’s exactly what I found with the prospectus, glad it’s not just me!
    It’s disappointing though, I’m sold on having a 10% allocation to commodities and I want the constant maturity/continuous rolling that UC15 provides but I can’t buy something which lists a 3% charge that I don’t understand!
    How did you contact UBS, I tried emailing but didn’t get a reply.

  • 68 The Accumulator August 16, 2024, 10:09 am

    I just emailed their head of ETF sales. FWIW, I believe that the charge does not apply but there are alternative ETFs that will do the job if you don’t want to take the risk.

Leave a Comment