Blog

Contents of this month’s post include:

• What is “Roll yield” and modelling its future impact on Treasury Bond ETF returns in a rising yield environment
• Update on my All Seasons Portfolio on eToro launched on 1 April 2021
• Monthly Update for December 2021 with a fresh set of charts
• Book tip: The Changing World Order: Why Nations Succeed and Fail by Ray Dalio (brief review and link at the bottom of the post)

Hi, and good to have you back for another article related to balanced portfolio investing!

I hope you have had a great start to 2022, and I mean that both personally and financially.

In this article, I will be sharing my last monthly update for 2021, and I have executed a rebalancing during December, as I had previously disclosed in last month’s post.

But before we get to that, I have another exciting text to share. Recently, a lot of discussions have been revolving around government bonds and whether they are still a sensible investment even in a balanced portfolio such as the All Seasons Portfolio, now that yields are rising and the West could be facing geopolitical uncertainty.

I wish I had a firm response on what to do with this, but this is something that I still ponder about.

But, when attempting to find answers on what to do with treasury bond investments, I began thinking about how roll yield could potentially be an important factor to consider when assessing bond returns. I will be explaining more in detail what that is further below, but I think you might find it interesting too how roll yield is likely to influence Long-Term Treasury Bond ETFs like the IS04/TLT (iShares $ 20+yr Treasury Bond ETF) in a scenario when rates and yields rise.

As I searched for more certainty what will happen with these investments, a key aspect of this research was to quantify the impact of roll yield. To achieve this, I modeled the returns by simulating 100 bond portfolios with similar structures as IS04 over a period with rising yields to see how they would behave, and compare that return with a portfolio that does not benefit from roll yield to see the difference. The results were quite clear actually, and I have been looking forward to sharing this with you in this article.

With this post, I am not attempting to convince you that investing in government bond is a good idea – I give no judgement in that. There are other factors than just roll yield to consider, and also the aspect that an isolated investment in bonds might be less than beneficial, but it might still make sense in a portfolio perspective.

Rather, I share my observations and findings from my research about roll yield as a phenomenon, and you can use that information as you wish in your analysis. My ambition is to be helpful in adding to your thought process and I hope we will use this as a collaborative process in improving our risk-adjusted returns.

Well, I will not spill all beans already now, but at least now you have an understanding about what I will be discussing below. And as always, please feel free to share your comments through any channel of communication you prefer.

---

When the coronavirus hit in March 2020, treasury bond yields hit their lowest levels since the second world war. This low water mark was not only caused by the Fed’s response to alleviate the impact that the virus and the lockdowns would have on the economy, but it has been part of a long-term declining trend.

Yields have been dropping steadily since they peaked in late 1981 when all of the 10Y, 20Y, and 30Y yields briefly hit above 15%. And more recently, spurred on by more dovish central banks around the world in the 2010s, yields have continued to fall in a fight against deflation that culminated when Covid-19 hit.

Perhaps the Covid-19 pandemic, and more importantly, the fight against it, happening concurrently with other macroeconomic factors, marks the end of the decline in yields. The currently rising inflation rate, which reached 7.0% year-over-year in the US in December and being at similarly high levels in Europe with 5.0% in the euro zone, together with stronger labor market participation, suggest that the Fed and other key central banks will begin raising yields again for the first time in years.

Not to only brush over “macroeconomic factors”, let’s stay on this for a moment. In the past years, Ray Dalio (Bridgewater Associates), a sort of a house god among risk parity investors, has criticized bond as an invest, as the risk/reward ratio has become rather uneven. With yields this low with zero-interest rate policy, there is almost only one direction where they could go, and that is up.

I say “almost”, as German investors were in the same predicament in 2014, being skeptical to the bund when they reached similar low yields as the American ones have today, not believing that they could fall any further. Rather famously, the long-term bund yield has seen been in deeply negative territory, at times trading at -0.5%.

But, as recently as 4 January 2022, Dalio published a new article on LinkedIn titled “The Changing World Order: The New Paradigm“, again presenting macroeconomical factors that should make investors more vary against American Treasury bonds. As Dalio is a successful hedge fund manager, turned economic historian and author, I find his articles thought-provoking and well worth the read.

Bonds in a diversified portfolio

As this is not meant to be a holistic article about bonds, and what you should replace them with in a portfolio if you want to avoid them, I will leave that discussion for another time.

Instead, I want to isolate this discussion to what will happen to your treasury bond investments if yields were to rise significantly from here. The fact is that treasury bonds form an important part of the All Seasons Portfolio strategy, as well as other similar strategies such as the Permanent Portfolio and Golden Butterfly, etc.

Government Bonds are one of few proper hedges against deflation and disinflation, as well as sudden negative changes to economic growth expectations. Few other asset classes provide that protection. Thus, there is still reasons to still include long-term treasury bond in a portfolio, even at low yield environments. Check also out this great article from Portfolio Charts discussing bond convexity and how bond price movements are amplified when the rates are negative.

Moreover, you may want to check out a fantastic article from AQR Capital Management further explaining what drives bond yields if you wish to read up on this more – I found this very helpful. One of the takeaways from that article that is worth repeating here is that a rise in yields for long-term bonds is not to be confused with a rise in the central bank funds rates. Yields longer out on the yield curve are affected by more factors, and the rates’ impact is only reflected as expectations of what the rates will be 20+ years into the future – not what it is today.

What can be expected of Treasury Bond ETFs going forward?

I will take the liberty to assume that you too are worried about the bond part of your portfolio in the current low-yield environment and what will happen to this part of your portfolio if yields were to rise.

The truth is that this has been bugging me too lately. I have wanted to find answers on how bad rising yields would be for my long-term treasury bond ETF (IS04 and the EUR-hedged versions IUSV/DTLE depending on stock exchange).

One could, at first glance, make the wrongful conclusion that rising yields will be as bad to the fund as the decline in yields has been good (1:1 retracement). Using TLT as an example (the American version of iShares $ 20+yr Treasury Bond UCITS ETF) as an example and looking at its returns since inception in 2002, it would mean that the +200% return would become a -66% loss if yields would return to the same levels as then.

Things are not that simple, however.

With such a simplified analysis, an important aspect of bond investments is forgotten, namely the impact of roll yield.

Finally, after a long while, we have arrived at the central theme of this article. The background was necessary though, as I believe it will be helpful for the further analysis.

But getting to the point, what is roll yield?

In summary, roll yield is the profit a bond investor makes from rolling a bond with shorter remaining time to maturity into a bond with linger remaining time to maturity.

When you invest in bond ETFs, they hold a basket of bonds, all with different initial yields and remaining time to maturity. To keep the desired remaining term, the fund will sell bonds when they reach a certain age and again buy bonds with longer remaining lifetime. For the IS04 ETF, that aimed remaining term is 20+ years.

To use a simple example though, let us consider an investment of only a US Treasury Bond with a term of 10 years. The yield of that one was 1.63% on 10 January when I last looked. But as the yield curve is (usually) upward sloping, that means that longer-term bonds have higher yields than shorter-term ones. This also means that if I hold my 10Y bond for 3 years, it will effectively become a 7-year bond and trade with a yield as other 7-year bonds. These bonds currently have a yield of 1.55%. Ceteris Paribus (“all else being equal”) by holding my 10Y bond for three years, I would automatically make a profit when its yields fall, as the price of a bond moves in the opposite direction of the yield.

As a bond is rolling down the yield curve, you make a profit, and as you keep rolling your holdings from shorter-term bonds to longer-term ones, you keep repeating this process over and over. With a basket of bonds (TLT currently holds 32 with remaining times to maturity ranging 18 to 29 years), that means that you would regularly be rolling your bonds, and thus be taking advantage of roll yield.

The size of the roll yield is explained by the spreads between the bonds. For reference, I have included here below a chart of the 30Y-20Y spread and 20Y-10Y spread from 1977 to 2021.

This means that at its core, the return profile for treasury bond ETFs is skewed to the right, as the roll yield will add to any returns derived from the changes in yields.

Additionally, as roll yields have amplified the returns for the past 40 years of falling yields, it is logical to draw the conclusion that they would be dampening losses if yields were to rise from today’s low levels.

 The impact of Roll Yield for Bond ETF performance

I have been interested in knowing how big the effect of roll yield would be. It is rather vague to just acknowledge that this phenomenon exists, so I have been interested in finding this out by quantifying the roll yields.

For this purpose, I have modelled this by simulating 100 bond portfolios. As I, unfortunately, am completely illiterate in any programming language, I have done this exercise in Excel. The results were not impacted by that, just that it took me so much longer in running all formulas.

A bit of background and assumptions for the simulations is in order here. For finding out how bond portfolios would behave with rising yields, we need a rate path to base our simulations on. As the exact path is not that important (remember, we are modelling just the roll yield component, not predicting exact returns), I have used the yield path from 1977 to 2021, but in reverse daily order (the right-hand side of the below chart, which is the forward-looking part).

In other words, the following rates are used:

Is it likely that the yield path will look this way? Couldn’t it be assumed that yields would rise much more sharply than what the reversed historical yield path would imply? Perhaps, but given the current debt-to-GDP ratio in the US, currently at around 120%, a sharp rise to 10% in rates would devastate the economy, which speaks against a very quick increase in yields.

Let’s also have a quick look at historical yield paths to see how these have behaved. After World War II, yields were also at low levels around the zero mark. From then, it took almost 40 years (35 to be exact) for yields to peak in 1981, and a majority of the upward movement happened in the last few years. While the below chart is only form 1962 to December 1981 (half the length of our above chart), the form of the two charts is similar, also if you just look at the second half of the simulated yield path. Hence, it is plausible that the scenario could play out, but here it is important to point out that the exact path of future yields has little connection to a simulation of roll yield.  

Then I needed to build my portfolios in a credible way, as returns will rely on which bonds are included (remaining time to maturity) as well as how many bonds you hold. To make the simulated portfolios similar to the IS04 (TLT) with around 32 holdings but which may vary from time to time, I have assigned a number of bonds per portfolio at random (using random.org) to each portfolio between 7 and 45 bonds.

Then, I devised the age of the bond at the start date of the simulation (3 January 2022) at random as well. Note also that the start date for each bond match the actual auctions on 15 February, 15 May, 15 August and 15 November, or the next business day. Then, when a bond’s remaining tenor reaches 15 years, it is rolled into a new bond with a remaining lifetime of between 20 and 30 years (selected at random). Note also that the portfolios do not take into account any interest payments, as it is only the prices of the bonds that are measured.

When the returns for each portfolio were generated, these were compared with a bond portfolio where the remaining lifetime was held constant through the whole exercise. The difference in performance between the average portfolio and the portfolio with constant remaining lifetime would indicate the magnitude of difference derived from roll yield.

Alright, alright, alright (quoting Matthew McConaughey), enough explanation about method already, let’s look at the results!

Let us start of with looking at how all our 100 portfolios did. As expected, the returns are negative, which is not surprising in a paradigm where yields rise from near-zero to closer to 16% over four decades. Later stages of this analysis will cover some more interesting findings around this though.

As you see, there is a bit of variability in performance, which depends on the bonds included in each portfolio. That is the importance of random assignment to make this simulation scientifically fair. The average of the portfolios (marked with green), saw a total return of -33.97 over the period, or a CAGR of -0.92%.

With a starting amount of $10,000, the end result was hence, on average, $6,602.75, with a standard deviation of $728.02 (or 7.28%). There were no outliers, as all portfolios returned withing 3 standard deviations from the mean, with the worst performer having an end-value at $4,437.87, and the best portfolio having $8,468.62. The end results follow a near-normal distribution.

Okay, but this does not on its own say much about roll yield; just that we may expect to lose money if yields were to rise.

Therefore, we will have a look at the comparison against the constant remaining lifetime portfolio that does not benefit from roll yield. In the chart below, the green line represents the average of the 100 portfolios, the gold line is the constant remaining lifetime portfolio, and the gray field measures the difference in percentage points (right axis) between the two lines.

Here, it becomes clear what role that roll yield plays, as it makes a significant difference in returns.

Rather than expecting a clear decline in portfolio value over the 40 year portfolio in the same manner as bonds have steadily risen in value since 1981 (more or less the gold line inverted, but with an added boost from roll yield) the green line trends sideways for long periods of time.

I would also like to comment that the difference is statistically significant at a 95% level of confidence when assuming the constant time to maturity portfolio has similar (percentual) standard deviation as the average portfolio, with a t-stat 17.97. The probability that this difference occurred at random is less than 0.0001.

Th most interesting aspects however, become visible when we compare the two portfolios (average of roll yield benefitting portfolios and the constant remaining lifetime portfolio) with the development of yields. Here I use the UST20Y as a proxy.

I find this to be the most important chart of this article. Instead of the almost perforce negative correlation with the 20Y yield, like the constant time to maturity fund has, the average portfolio benefiting from roll yield is much more stable. Sharp increases in yield will, however, also impact the value of the bond portfolio as the benefit of roll yield is a slower process that runs over years because it is the product of the spread size and the number of years it takes a bond to reach the remaining lifetime of a lower yielding bond.

Still, as becomes visible from the next chart where the green field have been replaced by the UST20Y yield being overlaid, even in a scenario where the 20Y yield rises from 2% to 8% over a longer period (between years 2028 and 2054 in the chart), the average portfolio trades completely sideways. You can see this clearly below with the same chart where this period has been shaded.

During this period, the investor could also take advantage of a rebalancing premium when the bonds portfolio runs between 80 and 100 (indexed) for long time, as the investor could buy at a relative low and sell at a relative high to rebalance against other assets such as stocks. Instead of the almost perforce negative correlation with the 20Y yield, like the constant time to maturity fund has, the average portfolio benefitting from roll yield is much more stable.

Conclusions and takeaways

From this exercise, a handful of conclusions can be drawn, which will be helpful for any investor contemplating the bond ETF’s place in a diversified portfolio:

Conclusion 1: Regardless which way yields (and rates) go from today’s levels, if they are climbing or turning negative, roll yield will add to the performance of a bond portfolio as long as the yield curve is not inverted. The return profile from bonds is thus, in general, skewed in favour of the bond holder.

Conclusion 2: The performance of a bond portfolio benefitting from roll yield, is about 16 percentage points better than a corresponding portfolio with constant remaining lifetime over the simulation. This difference is statistically proven.

Conclusion 3: In a scenario where yields rise slowly and steadily, like between 2034 and 2052 in the simulation when yields rose from 2% to 8%, a bond ETF will trade sideways.

Conclusion 4: In such a scenario when bonds trade sideways, it retains certain volatility from which the risk parity investor can reap rebalancing premiums, as the low correlation with for example stocks would persist.

Conclusion 5: Roll yield does not protect against quick jumps in yields, as this phenomenon is slower and works over many years.

Conclusion 6: Roll yield only adds to bond portfolio performance when the yield spread is positive, i.e., when subtracting the yield of a shorter-term bond from that of a longer-term bond. If, for example, the 30Y-20Y spread is negative, as it has been during long periods of times (and is currently), it may be beneficial to wait with rolling the bond to a longer one. This poses an interesting topic to dissect further when constructing a bond investment strategy.

Conclusion 7: Roll yield is applicable to bonds issued by all governments, why you should consider this also when investing in for example German bunds, UK gilts or any other government bond.

Conclusion 8: While there are several valid arguments against investing in bonds at this time, which remain a concern for me as well, roll yield speaks in favor of this asset class, and it is shown here that simply an increase of yields is not as strong of an argument as may have been assumed.

Conclusion 9: Even when yields rise sharply (see years 2058-2060 in the chart), the decline in the value of bond portfolios is subdued thanks to the roll yield, and the drawdown is smaller than if roll yield was not a thing.

Summary and ending comments

While this has only been a simulation assuming a reversal of the historical yield path, it shows what difference roll yield makes to a bond portfolio in adding to profits and dampening losses. The return profile therefore becomes skewed in favor of the investor. It is of course impossible to predict how the yield path will come about, but that is less relevant in this illustration.

A caveat is of course that there may be other aspect that are making bonds an unattractive investment, for example the factors promoted by Ray Dalio in his LinkedIn posts or his new book The Changing World Order, which discusses America’s state in particular, which is why he suggests avoiding investing in bonds issued by the US Treasury.

Few other asset classes offer the same protection against disinflationary environments and growth shocks as treasury bonds. At the same time, there are challenges associated with investing in bonds at these yields and in the current macroeconomic environment, I acknowledge that, and I am concerned about that. But an important part in making the investment decision is to quantify what we can to start the analysis, and I find that roll yield is an important aspect to consider when thinking about bonds as an investment.

I hope you found this article as interesting as I found working on this research, and that you got some valuable takeaways from it. Let’s continue the discussion in the comment section or by any other means of communication you prefer.

---

December 2021 Portfolio Update

December all in all gave som slight positive return, but my All Seasons Portfolios lagged the stock markets this month.

In December, global equities rose despite rapidly rising Omicron cases and higher inflation. Investors were not particularly worried about the surge in covid cases, as hospitalisations and deaths remained at low levels.

Fixed income declined this month, as Fed announced an acceleration on tapering, and this movement downwards in treasury bonds continued through January when the minutes from the Fed’s December meeting were published. Hence, the gap between the S&P 500 and a 60/40 Portfolio was wider this month.

Gold prices traded sideways over the month, closing up 0.1% in USD, failing to breach the important USD 1,830/oz.

Electricity prices spiked substantially in December, pushing up prices in energy commodities. Oil traded well, with BRENT rising about 7% over the month alone to $78.6/barrel. This strong trend has also continued into 2022. Broad commodity indices also traded upwards with the Bloomberg Commodity Index rising 4.17%, and CRB/Refinitive Commodity Index rising 6.4%.

In alternatives, crypto currencies have been in a slump lately and trading steadily downwards for quite some time, having lost $10,000 from $57k on 30 november to about $46k on New Years Eve. With the strong equity market, VIX also was under pressure in general, but showed some trading opportunities that I was able to capture in my eToro portfolio where I trade these this asset more frequently to add a bit more overall stability to the portfolio performance.

Speaking of eToro, the month was not the best one, as my All Seasons Portfolio saw a slight decline of -0.40%. Overall volatility remains at very much acceptable levels, and my first 9 months on the platform saw an aggregate return of +13.40%, which I am very much pleased with.

January has started in a similar manner, but still ahead of stocks, showing that the expected returns are much more stable.

If you are already on eToro, make sure to follow me there too, as I from time to time share brief updates there directly about that particular portfolio. The updates I share on this blog will however remain deeper and more insightful.

Follow me there by finding user Allseasonsport. And feel free to copy my portfolio there with a small amount of your portfolio if you want a more hands-off approach to risk parity investing. I do all my trading there in a systematic and rule-based manner, and already have 16 copiers at the time of writing. I very much like this copy investing functionality, as it makes it easier to follow other people’s strategies, while the investors like myself that are copied have skin in the game as all trades are done with my own money.

As for my “regular” All Seasons Portfolio, the one I am trading on DEGIRO, also saw negative return over the month.

In December I performed an annual rebalancing, going into a new uncertain year with fresh allocations. This is quite reflected in the current allocation.

However, with the uncertainty around rising yields and rising inflation, I have not rebalanced all the way back to the aimed allocation with Long-Term Treasury Bonds, Gold and Commodities. This is in line with what I wrote about in my last monthly update about Strategic Rebalancing, and has turned out to have been a decent choice so far in January.

• 

• 

As I mentioned already, equities saw a good month, which is why my portfolio has been lagging the S&P 500 a bit at late.

There seems to be something odd in my data stream for S&P 500 though, as volatility is through the roof as you’ll see in the table to the right. I’ll see if I can have this fixed for next month.

More importantly, my VAR and volatility levels remain spot on at good levels, and the absolute returns have been great the past 12 months, with a Sharpe ratio I am quite proud of. This is why I am comfortable with levering my portfolio 1.3x on eToro, as the vol is still less than for stocks, but where my returns can get a boost.

• 

• 

Looking at individual assets, stocks and commodities were the main drivers on the positive side, fixed income trading just slightly down, and then alts being down materially. But as alts have a small portion of my capital allocation, this affects my wealth only on the margin.

Adding also the 3-month chart here below with most assets being fairly grouped upwards.

Lastly, as usual, here is the table of my ETFs and the changes laid out in table form.

ETF	Name	Asset Class	2021-11-30	2021-12-31	Change (portfolio)	Change (ETF)
UIMB	UBS LFS Bloomberg TIPS 10+ UCI ETF(USD)Ad	TIPS	€694.45	€564.20	-18.76%	-0.19%
DTLE	iShares $ Treasury Bd 20+yr UCITS ETF EUR Hgd Dist	Long-Term Government Bonds	€843.92	€939.84	11.37%	-2.55%
IGLE	iShares Global Govt Bond UCITS ETF EUR Hedged Dist	Long-Term Government Bonds	€637.26	€632.10	-0.81%	-0.81%
M9SA	Market Access Rogers Int Com Index UCITS ETF	Commodities	€514.74	€417.78	-18.84%	3.71%
4GLD	Xetra-Gold	Gold	€456.16	€410.59	-9.99%	1.26%
FLXG	Franklin LibertyQ Global Equity SRI UCITS ETF	Stocks	€769.84	€799.56	3.86%	3.86%
JPGL	JPM Global Equity Multi-Factor UCITS ETF - USD acc	Stocks	€823.14	€799.83	-2.83%	4.64%
VOOL	Lyxor S&P 500 VIX Futures Enhcd Roll UCITS ETF A	VIX	€100.34	€86.42	-13.87%	-13.87%
Currency:BTCUSD	Bitcoin XBT	Crypto	€256.25	€208.10	-18.79%	-18.79%
Added cash			€0.00	€0.00
Total			€5,096.10	€4,858.41	-4.66%

---

Thank you yet again for following my blog about risk parity investing and the All Seasons Portfolio. If you haven’t done so already, make sure to subscribe to the newsletter via the form in the page footer, and to drop any comments you may have on the content with the comment section or via email to nicholas@allseasonsportfolio.eu. The greatest value I have received from upkeeping this blog is the fantastic conversations with great people, such as yourselves, about ideas on investing and strategies. Thanks for that!

Remember also to check out the Resources page which I will be filling up with useful tools and resources for managing an All Seasons Portfolio. If you have any suggestions for any particular spreadsheet or tool that you are after, let me know in the comments below and I will see what I can do to have it included.

If you wish to support me in alternative ways without donations, you can share the post through social media or with your friends who should invest more responsibly. Alternatively, occasionally click any ads or access Amazon via the links included in the book tip section below any time you are making any purchase from their platform (it does not necessarily have to be the book I am linking to).

We’ll catch up soon for the October update!
Nicholas Ahonen

---

Want to show your support and help financing the maintenance of the blog through a one-time donation? Leave a sum of your choice, like $5, to the tip jar via Paypal (common credit cards supported). Any contributions are highly appreciated!

---

Book tip: The Changing World Order: Why Nations Succeed or Fail by Ray Dalio

NEW YORK TIMES BESTSELLER

“A provocative read…There are few tomes that coherently map such broad economic histories as well as Mr. Dalio’s. Perhaps more unusually, Mr. Dalio has managed to identify metrics from that history that can be applied to understand today.” – Andrew Ross Sorkin, The New York Times

From legendary investor Ray Dalio, author of the #1 New York Times bestseller Principles, who has spent half a century studying global economies and markets, Principles for Dealing with the Changing World Order examines history’s most turbulent economic and political periods to reveal why the times ahead will likely be radically different from those we’ve experienced in our lifetimes – and to offer practical advice on how to navigate them well.

A few years ago, Ray Dalio noticed a confluence of political and economic conditions he hadn’t encountered before. They included huge debts and zero or near-zero interest rates that led to massive printing of money in the world’s three major reserve currencies; big political and social conflicts within countries, especially the US, due to the largest wealth, political, and values disparities in more than 100 years; and the rising of a world power (China) to challenge the existing world power (US) and the existing world order. The last time that this confluence occurred was between 1930 and 1945. This realization sent Dalio on a search for the repeating patterns and cause/effect relationships underlying all major changes in wealth and power over the last 500 years.

The topics of this book are incredibly timely, insightful, and thought provoking, as Dalio lays out studies of world’s leading empires through history like the Dutch, and the British, and their arcs of power, before they all eventually were superseded by another. Interestingly the phases for each such arc seems to have been repeated for centuries, and the current dominant power – the US – appears to be following the same course of a “Big Cycle”.

He reveals the timeless and universal forces behind these shifts and uses them to look into the future, offering practical principles for positioning oneself for what’s ahead when the next probable leading power – China – challenges the current paradigm as it is in the early phases of the arc.

This is one great read, and a book you should pick up to better understand big cycles, as these aspects are important to implement into a well-diversified portfolio that should withstand the test of time.

Or check out other great books on the topic on the Book recommendation page.

Buy it today on Amazon (affiliate link):

Buy it on Amazon.com

---