Institutional investors with limited risk capacity must include risk-free investments in their allocation, even if this involves costs. An interest rate normalization is not in sight. All the more reason for them to improve their returns through effective diversification and optimized risk management.
The current low interest rates correspond to a
long-term trend of falling real and nominal interest rates.
Central banks are not to blame for the low yields, rather
they have reinforced and accelerated a long term trend.
Even if monetary policy normalizes, interest rates -
and thus yields - are likely to remain below the long-term average.
Does it still make sense to invest in government bonds
or low-risk bonds with negative yields? The answer is “YES”. First-class bonds
diversify and stabilize the portfolio. They belong in the portfolio, even if
this involves paying a premium.
Investors should not compensate for the low/negative
bond yields with additional risk or more complexity. It is better to improve
the efficiency of the portfolio through optimized risk management.
An efficiently diversified equity portfolio allows for
a higher equity exposure without exceeding the risk budget. It thus also
delivers a higher expected return.
Central bankers are not superheroes
As a high school
student I was fascinated by the central banks and their power to print money.
In my naive vision of the economy, they would be able to solve the world's
problems with their "seigniorage power". Later, at university,
disillusionment followed: I realized that my supposed superheroes could neither
do magic nor print money at will without risking doing more harm than good. The
greatest risk would be to lose their highest attribute, namely their
credibility. I have also learned that the interest rate cannot drop below zero:
Why would anyone give up consuming today and instead lend money to consume
tomorrow, if tomorrow they will receive less than today?
... or are they?
Today even orthodox
principles are crumbling: The central banks' money printing machines are
running relentlessly, leaving a quarter of the world economy (measured by GDP)
with a negative interest rate. Was my once naive view of the role of central
banks correct after all? Are they superheroes after all? The economy is doing
well, unemployment is relatively low, there are no inflationary pressures, the
financial markets are delivering substantial positive returns and the
credibility of the central banks is not being put into question (yet). Do we
really have to throw out the classics among economic textbooks? There is a
lively debate in the media right now between those who are concerned about the
nonchalant attitude of central banks and those who see the banknote press as
the solution to the world's economic problems. Among the latter are the advocates
of Modern Monetary Theory. Regardless of who is
right and who is wrong: If the central banks stop the printing press and bring
the real interest rate back to at least zero, most investors, especially
institutional investors, will have more sleepless nights.
Investment emergency postponed
Supported by the
ultra-expansive monetary policy, the financial markets have continued to rise
unabated over the last ten years. In 2019, most stock markets recorded
historically high returns. At the same time, the term pénurie de placements or Anlagenotstand, in English “investment
plight” has been circulating in the Swiss media for about five years and the
term is ubiquitous at every institutional investor conference. Given the juicy
returns of the last five years, one may ask: What pénurie de placements?
Have the investors who
have had to deal with the anomaly of a negative risk-free interest rate for
almost five years learned how to deal with it, despite the fact that no
textbook provides for such a possibility? Negative interest rates affect
Switzerland as well as the Eurozone and - for some time now - Japan. The USA is
still recording slightly positive interest rates.
The good portfolio
results were possible because of capitalized future returns, i.e. they are already
included in today's prices. However, this means that the returns expected in
the future will be correspondingly lower. Indeed forward looking, investors are
facing higher challenges than in the past years.
Replacing unrewarding investments?
For a long time it was
important to avoid "Japanization" - an economy with negative or
near-zero inflation, permanently low interest rates and low growth. Well-known
analysts considered this scenario impossible for the European or North American
economy. Today, 15 years later and with negative interest rates, we are in an
even more difficult situation than the Japanese asset managers were back then.
The question is: Is it
wise to invest in assets with a zero or even negative return? Shouldn't such
unrewarding bonds simply be removed from the asset allocation? There is a
strong temptation to replace them with substitutes that have a positive return
and a recurring income. Typical candidates for substitution are real estate and
corporate or high-yield bonds from debtors with poor credit ratings. In
addition, alternative investments such as infrastructure investments, private
loans or private equity are also considered substitutes. What most of them have
in common is that their value is based on expertise and appraisal value - and
not on a daily market price. The danger of forgetting the risks - implicit or
explicit - is huge.
Learning from Japan
We advise investors not
to switch negative risk-free returns for greater complexity and risk, as we
explained in last year's research note. Rather, we recommend
sticking to the risk budget until the possibilities for risk optimization have
been exhausted. Japanese portfolio managers stoically continue to buy Japanese
government bonds, even if they have a negative return. So what is the
motivation for investing in these assets? The answer is very simple: consistent
management of portfolio risk and/or decorrelation.
Decorrelation and risk dilution
A key feature of
investments in risk-free government bonds for a multi-asset portfolio is risk
mitigation. Apart from the decorrelation effect, which is also common in other
asset classes, the so-called risk dilution effect also contributes to risk
reduction, and in a significant way. It arises when we combine high-risk
investments with bonds that have little or no risk (measured by standard
First-class (risk-free) government bonds have the lowest volatility and thus
the highest risk dilution effect. They are the only investments that guarantee (nominal)
capital repayment at maturity. At the same time, the low or even negative
correlation to equities, as we have experienced over the last 20 years, offers
a further contribution to risk reduction via the decorrelation effect. Especially
in times of crisis, risk-free government bonds continue to be a safe haven.
They are therefore important for stabilizing the portfolio. Figure 1 shows that
the correlation between government bonds and equities has remained relatively
low over time and their volatility ratio (a measure of risk dilution) stable.
Other asset classes, such as commodities, may also provide a
decorrelation effect. Here, however, the risk dilution effect is almost zero,
as commodities have a similar, if not higher, volatility than equities.
effect is based on the uncorrelated yield movement of bonds relative to other
assets. The price changes of bonds are determined by interest rate sensitivity,
which depends primarily on the maturity and is measured by their duration. The
higher the interest rate sensitivity (duration), the higher the diversification
effect. Whereby a longer duration also implies a higher price variability and a
slightly higher standard deviation compared to bonds with a short duration.
Figure 2 illustrates
the decorrelation effect and risk dilution for a simple portfolio of bonds and equities,
assuming a volatility of 25% for equities and 5% for bonds (duration of 5).
With correlation 1 we only have a risk dilution effect. With a lower
correlation, the decorrelation effect increases until a hedging effect
(correlation -1) is reached.
The attractiveness of
high-quality corporate or government bonds in a multi-asset portfolio manifests
itself most clearly in times of crisis. Then, the guaranteed repayment of the (nominal)
capital becomes more important than the return. Under such circumstances,
first-class bonds show an even lower correlation (direction -1) than the one
they have on average and offer partial protection against losses, especially in
the case of bonds with a long maturity. This is an important contribution to
portfolio stabilization, as can be seen in the table (Figure 3): Over the last
30 years, the global government bond index hedged in CHF performed best when
the equity markets (MSCI World Index in CHF) experienced the biggest drawdown.
Risk reduction has a price - and a value
So far so good. We have
just refreshed the rationale for investing in government bonds which, as
risk-free investments, guarantee repayment of the nominal value at maturity. In
almost a quarter of all markets, such an investment costs money. For CHF investors
in particular, risk reduction through bonds in the portfolio has the highest price compared to other reference currency. All
the more reason for them to focus on optimal portfolio diversification. This is
actually nothing new, but is based on modern portfolio theory (Harry
Markowitz). Instead of replicating a capital-weighted equity index (indexing),
we recommend optimizing the equity portfolio. Our risk-based portfolio
optimization reduces the risk by 20-30% and thereby allows for a higher
proportion of equities at the same risk level (see Figure 4). The improved
diversification and higher efficiency ultimately lead to a higher expected
In summary, efficient
strategies such as OLZ Optimization improve returns without compromising
complexity, cost, liquidity risk or governance. In the current market
situation, effective risk management is therefore more important than ever.
Outlook: Declining expected returns and efficiency
The global decline in
interest rates and credit spreads over the past decade has had a significant
impact on expected returns and the efficiency frontier. Even in the US, where
risk-free government bonds still have a slightly positive return, the efficient
frontier has shifted downward and flattened. Figure 5 compares the long-term
historical efficient frontier (30 years) for USD-based investments with the one
over the last 10 years, which is higher. Of particular concern is the ex-ante
efficient frontier for the next 10 years, which is significantly lower than the
historical one and extremely flat.
This means that the effect
of the pénurie de placements (investment
plight) has not materialized yet, because
it has been postponed into the future. In Figure 6 we have calculated the
efficient frontier for CHF-based investments and compared the last 10
years to expectations for the next decade. Again, expectations are below the
status quo. Overall, falling expected returns lead to a lower and relatively
flat efficient frontier.
Conclusion: Bonds for stabilization and efficient
equity portfolios for higher returns
Since most investors
have a limited risk budget, they must include risk-free investments in their
allocation, even if this involves costs. Investing in bonds acts like an
insurance policy, where you pay an annual premium to reduce risks. If diversification
and risk reduction is desirable and investors are willing to pay for it, then
they should give due attention to risk management throughout the investment
process. Efficient risk-optimised equity portfolios will play an important role
over the next decade, and in the long term they will help to generate higher
Low and even negative
returns make risk management a central task of any investment strategy. The
goal should always be to build an efficient portfolio given a specific risk
Central banks - the real "bad guys"?
Negative nominal interest rates and ballooned balance sheets are lasting, and the central banks have communicated this more or less explicitly. The extent to which the financial markets are dependent on the central banks medicine was experienced in November and December 2018. The panic-like market correction immediately led to a return to the proven expansive monetary policy that has characterized the last 10 years.
It is not bold to say that interest rates will normalize anytime soon. Nevertheless, the question that arises is what a normal interest rate level is.
What is a normal interest rate level?
What can be considered a normal interest rate? A very informative and well documented study was recently presented by Prof. Schmelzing of Harvard University. Using a fabulous collection of data spanning 700 years, he shows that real and nominal interest rates do not fluctuate around a constant "normal" level, but rather fall continuously. The last 30 years, which have been characterized by falling nominal and real interest rates, are therefore not an abnormality but a convergence towards the historical trend of falling average interest rates.
One conclusion is that negative real interest rates are not unusual. Over the 700 years that have been analyzed, the share of countries/economies (measured against GDP) with negative real interest rates has increased from an average of 15% to about 30%. For the coming decades, Prof. Schmelzing dares to predict that real yields could become negative. Apart from the strong implication of this statement, the study shows that historical average nominal and real yields are not a good reference. Rather, we have to assume that average values will decline.
Furthermore, the study comes to the conclusion that central banks have at best accelerated, but certainly not triggered, a long term interest rate dynamic in developed economies. We would have had to deal with falling or near-zero real interest rates anyway - even without an expansive monetary policy. From this perspective, "central bank bashing" seems exaggerated.
 Paul Schmelzing, “Eight centuries of global real interest rates, R-G, and the ‘suprasecular’ decline, 1311-2018”, SSRN, November 2019.
The study covers advanced Western economies/countries in historical
comparison. The author adds up the GDP of countries with negative real
returns and then calculates the percentage in relation to the total GDP
of his sample.
Wie wird sich das How will the "normal" interest rate level develop?
Central banks can do a lot, but they are by no means the "masters of the universe" as the last decade has suggested. The unusual monetary policy of the last 10 years has reinforced a fundamental historical dynamic of the market towards lower nominal and real interest rates. The new normal interest rate level (where an economy is in balance between growth, inflation and employment) is below the historical average. Therefore, a normalization of monetary policy would only lead to a moderate increase in interest rates. Contrary to what many have feared, this would mean a lower potential loss for interest-sensitive assets.
 OLZ Research Note- January 2019: Efficiency instead of Complexity
 Leibowitz, M. and A. Bova; The duration decision in diversified
portfolios; Morgan Stanley Research, 2012. See also: Norges Bank Investment
Management: “Risk and return of different asset allocation”, discussion note;
 It is interesting to note that commodities do not yield dividends and
interest or other income and in addition bear a negative return in the form of
 We have considered the following
asset classes: Cash (5%-30%), Bonds CHF (10%-50%), High Yields (0%-15%), Real
Estate (20%-35%), Swiss Equities (10%-30%), World Equities (5%-30%), Emerging
Markets Equities (3%-20%), with a max. 80% equity investment. The expected
returns are derived from: Invesco Investment Solutions, "2020 Invesco Long
Term Capital Market Assumptions - Swiss Francs (CHF)", October 2019. The
expected variance-covariance matrix is equated with the historical one.