### Bonds vs Stocks – a data-driven approach to long-term asset allocation in today’s high-rate, strong-equity environment

…or simply put, our models are showing that at this point in the market cycle, bonds and commodities look fairly superior to equities, with a particular emphasis on global bonds.

Global equities, and in particular, US Equities, have now had a very extended period of unprecedented returns. At the same time, the Fixed-Income sector has performed, well, awfully. It’s human nature to look for patterns - we look at this past performance and assume it will continue indefinitely, but this can also be potentially foolish at best, and dangerous at worst.

A more rational, numerical approach is to look for estimators of future returns and risk, and to attempt to use those predictors to choose portfolio asset allocations that reflect those expectations. The biggest challenge with this approach is that it’s very difficult to estimate returns of asset classes over short periods of time – months and quarters, while we believe that estimators of long-term returns can be quite effective. Long-term returns are useful to many real investors who have these long timeframes in mind as they save for retirement or inheritance, but long-term timeframes are of little use to most advisors and money-managers who rely on a steady stream of monthly and quarterly returns to support their businesses.

At FinMason, we have developed a set of models of estimators of 10-year future returns of 16 global macro market factors, which historically have been quite strong estimators. For example, for US Equities, we use a modified Case-Schiller model that uses long-term Price/Earnings ratios. We then use these factor estimates to estimate the forward returns and risk of any given portfolio of funds and equities.

The following table shows FinMason’s Capital Market Assumptions, based on these models, compared with historical returns and risk levels.

**Methodology:**

We used our models for a typical set of 19 large ETFs representing the major asset-classes and sub-asset-classes, and estimated their forward 10-year returns, volatilities, and covariances.

We then asked FinMason's optimization engine to answer the following questions:

**Q1: **Based on these estimates, is there an optimal allocation of these ETFs? i.e. what portfolio of ETFs gives the largest historical Sharpe Ratio? (Sharpe Ratio is a standard measure of returns relative to the risk in the portfolio). We first constrained the optimization to limit allocation to any one ETF to a maximum of 50% of the portfolio.

A: A portfolio of 50% Global Bonds, 28% US Intermediate Term Govt Bonds, 10% Emerging Market Equities and 12% Commodities, had the highest estimated Sharpe Ratio of 0.54, and an estimated return of 5.6%. Note that this is higher than the S&P500 fund at 5.0%, but with a lot less risk – the volatility of 4.4% is less than 1/3 of the S&P500 which has a volatility of 15.9%. The downside is that this is a pretty concentrated portfolio. Note that of the 13 equity ETFs offered to the engine, it only chose a small allocation to one fund - emerging markets. The explanation here is that, even though some equity classes have higher expected returns than the bond ETFs, the added risk involved is not worth it from a mathematical perspective.

**Q2.** What if we wanted more diversification, limiting the maximum allocation to 25%, what is the optimal allocation then?

A: As requested, the engine reduced the allocations to Global Bonds to 25%. It added a 25% allocation to US Intermediate Corporate Bonds, but this was mostly offset by removing the 28% allocation to Intermediate US Government Bonds. The reduction in global bonds was reassigned to equities – raising the Emerging Markets allocation to the maximum 25% and the engine also added a small 2% allocation to Technology Equities. It raised the allocation to Commodities to 25%. This resulted in a more balanced allocation of a total of 50% to bonds, 27% to equities, and 23% to commodities. The expected return increased to 7.0%, with the expected volatility increasing to 9.3% - still less than 2/3 of the S&P500. The Sharpe ratio fell to 0.40.

We are not suggesting that any of these portfolios are better than any others, but this analysis indicates how attractive bond funds are now vs most equities, within a reasonable portfolio allocation decision. We had previously run this analysis in early October, before the large rally in the bond markets, but despite that rally, the engine is still making large allocations to bonds when trying to make a long-run allocation.

The table below shows the two portfolios described above vs a 100% S&P500 fund allocation. It is easy to see how investors might be overly biased towards equities if they are just looking at historical returns. We recommend including a view of forward-looking returns, when thinking about asset allocation for the long run.

In addition to traditional measures of risk such as volatility and drawdown, we also show the FinScore™ of each portfolio.

**FinScore™** is FinMason's proprietary 1-100 risk score. It is an easy way for wealth managers and their clients to understand how risky a portfolio is. It is measured by the portfolio's sensitivity to FinMason's global 16-factor macro model. It has the benefits of representing long-term risk and is very stable over time and through changing market conditions. It is also easy for wealth managers to understand, because it is benchmarked to the five standard risk-target categories:

For more information on how FinMason can help with portfolio allocation questions or other investment analytics and data needs, please visit www.finmason.com or email us at [email protected].

* FinScore™ is the FinMason Portfolio Risk Score, based on its FinMason Macro Factor Model. Scores range from 0 to 100, and are benchmarked to 10=Conservative, 30=Moderately Conservative, 50=Moderate, 70=Moderately Aggressive, 90=Aggressive.

** 10-year forward expected return based on FinMason's 16-factor macro model. Results are as of Dec 31, 2023

*** Volatility is an industry standard measure of risk. The easy way to understand volatility of, for example 12%, is to realize that the one-year returns of the portfolio are expected to stay between +12% and –12% above and below the expected return, two-thirds of the time (i.e. on average, two years out of three will be in this range). The one-year returns are expected to stay within double this range, +24% to -24% above and below the expected return, 86% of the time, and within triple this range, +36% to -36% above and below the expected return, 99.7% of the time, assuming what we call a “normal distribution” of returns.

**About FinMason**

FinMason was founded by experienced industry insiders and leading technologists who have managed institutional portfolios, built and managed performance, risk, and analytics systems for large institutions, and built and sold large technology companies. FinMason solves the two largest hurdles in investment analysis – wrangling market data and calculating analytics at scale. Via FinMason’s cloud-native API the company provides a lightning-fast, customizable, calculation engine to accelerate any wealth technology build out. For more information about FinMason, visit www.finmason.com or email [email protected].

**Important Note**

Please note that the above estimates are based on a set of mathematical models. All investments are subject to risk and may lost money. FinMason does not claim to be able to accurately estimate forward returns of the market and FinMason is not an Investment Advisor. This analysis is presented for informational purposes only. Investors should seek the advice of a Registered Investment Advisor before making investment decisions.