In Part I of this series we looked at building a portfolio to meet minimum returns expectations. If you’re looking for 10%-12% returns per year, that’s difficult to achieve over the long-run just by picking assets, so generally some leverage has to employed. That post explained that an approach where the portfolio is diversified among multiple asset classes and weighted appropriately with moderate leverage can be safer than a highly concentrated portfolio with no leverage. This is because concentrated exposure increases risk exponentially, not linearly, as displayed in the graphs below. That post also argued that the approach works best when the assets in the portfolio are uncorrelated. The lower the correlation, the better the reward-to-risk ratio in a portfolio. This was demonstrated in the scenario of having a bunch of equal-returning, equal-risk assets and how reward-to-risk would be impacted based on their correlations. 75% correlation 50% correlation 25% correlation No correlation In terms of “uncorrelated,” that doesn’t mean in the traditional sense that asset A and asset B are X% correlated over the past Y number of years. Rather, correlation should be conceptually modeled in terms of what fundamentally drives movements in different assets instead of simply backtesting what happened. What occurred in the past can provide an empirical means of testing how something worked in practice through different economic and market environments. But the future can deviate from the past and the principles that you base your decisions on need to be deeply logical or else the results are unlikely to be optimal. In other words, why do investors buy and sell assets when they do? When thought of from this perspective, it is easier to understand why they might move together, move apart, or have no covariance. If you can figure out that right mix between credit, currencies, equities, commodities, and other viable asset classes and sub-markets, you can essentially be “market neutral.” Namely, you can be agnostic as to what growth, inflation, interest rates, and whatever else does. It’s vital to pay special attention to risk because everybody’s favorite asset class, whether that’s equities, credit, commodities, will decline by some 50%-75% if they hold a portfolio over some 50+ years. The larger the drawdown, the larger the gain required to get back up to your breakeven mark. This is also an exponential relationship, not a linear one. If you suffer a 10% drawdown, you need only an 11% gain to get back up to where you were, which is very manageable. However, if your drawdown is 5x that amount (-50%), then your required gain now increases almost 10x (100%). If you’re drawn down 80%, which is common in portfolios concentrated in one or a few investments that sour, building that back up could literally take decades, if ever. This can be particularly disastrous in cases for those getting late in their careers and have already earned most of the money they’ll ever make. In Part II of this series, we looked at approximations of the risk and return of individual asset classes and how to leverage these in certain ways to get your expected returns. Then, building off part I, how to weight these positions in such a way that your return is maximized at the lowest possible risk. Or put differently, gaining the maximum return per each unit of risk. And how this portfolio can be superior in risk-adjusted terms relative to common configurations, such as 100% stocks or the “50/50” (50% stocks, 50% bonds) portfolio. _____ Beyond Part I and II The rest of this post (“part III”), builds off the same principles, but focuses on individual assets and pinpointed sources of alpha, rather than focusing strictly on the asset class level. For example, let’s say you’ve determined your optimal (“market neutral”) asset mix to be 20% equities, 65% bonds, 5% commodities, and 10% currencies and rates. You could construct a decent portfolio modeling this with just a few exchange-traded funds (ETFs), bond investments, and exposure to the futures and FX markets. (There are ETFs for these markets as well, but it’s generally not the most efficient way to gain that exposure.) This might give you the basic underlying structure of what you want, but there would be little direct targeting of specific alphas. For example, let’s say for your own portfolio you want to be allocated about 40% in equities, give-or-take, based on the risk associated with the specific sources of alpha that you choose. You don’t want to have generic exposure to US equities (e.g., $SPY) because you feel you can improve on that by investing in companies that you believe will outperform the index. In other words, you’re looking to “overlay” alpha on the beta by choosing sources of alpha that you believe are best based on your own personal research efforts. For example, let’s say you like e-commerce and fintech. For those using a tool like WhoTrades Live, selecting individual portfolios to invest in is a matter of screening for various themes of interest and successful traders. You can either do this directly through the live.whotrades.com URL, or on the WhoTrades site by typing in tickers at the top and looking at portfolios that invest in a particular stock of interest. Take Alibaba ( $BABA ) for example. This particular portfolio contains various e-commerce and fintech names, such as $AMZN, $BABA, $SQ, and other names in higher-growth verticals such as $NVDA, $DBX, $NFLX, and $SPOT. (Source: WhoTrades Live Portfolio Page) If you’re looking for financials only, you can find something like that, as well. There are thousands of individual traders and portfolios on the platform. (Source: WhoTrades Live Portfolio Page) There is no limit to the number of portfolios you can invest in, and each one can be considered its own separate return stream, and combined into one portfolio. And even if you were to get the alphas wrong – or if the assets you think will outperform your benchmark, on net, do not – if you have balance in the portfolio, the underlying “market neutral” beta structure will mute the impact. The only time financial assets underperform cash for any material duration is during economic depressions, and in these cases central banks and governments have strong incentives to do what they can to get the economy moving on the right course again. As such, these tend to not last for elongated durations when the causes are properly identified and the policymakers have the authority and appropriate plan to resolve the problem. So in the case of the first approach, where you’re balancing the portfolio in a market-neutral format such that you’re not biased to specific growth, inflation, and interest rate outcomes, your allocation would look something like this: The above approach will yield a strong risk-adjusted return over the long-run with relatively limited drawdowns (depending on how the assets are leveraged). But the alphas will be poorly diversified and tied to the underlying asset classes rather than on the basis of what alpha streams are best. In the case of the improved approach, you’re maintaining the same fundamental structure of the portfolio, but choosing individual alpha returns streams to enhance what you’re doing in the first approach. The individual alpha returns streams can be individual assets or individual managers’ portfolios that are weighted in accordance with your reward-to-risk preferences. For example: The asset class allocation is the same between each, but example alpha sources are singled out in the second approach. In most portfolios, you’ll find that large amounts of exposure are in US equities. Or more generally, most exposure is concentrated in the easiest-access markets for investors, which are typically their domestic equity markets. This means that a large amount of both the alpha and beta-related returns will be associated with and concentrated in this market. On top of this, many investors compete exclusively or near-exclusively in the equity markets and ignore other markets available to them – even though, for instance, the global bond market is some 2.5x deeper than the global equity market in terms of monetary value. Accordingly, generating alpha from the US equity market is difficult. And the alphas, along with the beta-linked return, will be non-optimally diversified because so much of the portfolios returns are reliant on a particular asset class and thus a particular set of fundamental drivers – namely, strong economic growth and a tame inflation and interest rate environment. When the alphas are balanced appropriately and less correlated to each other by coming from diverse sources and markets, the concepts are the same as discussed in part I – this can lead to a much higher return for each unit of risk. Therefore, higher returns can be achieved for similar risk, or similar returns can be obtained for less risk. But this, of course, assumes that the alphas are chosen well. If they are not, the “alpha overlay” strategy will be expected to underperform the standard beta portfolio. Conclusion Following these concepts, one can enhance their returns by balancing their beta and alpha exposures in a market-neutral format such that the portfolio is largely impartial to moves in economic growth, inflation, and interest rates, as the most basic and overarching macroeconomic variables. This approach will improve risk-adjusted gains over simply investing in one asset class and come with lower drawdowns and higher return per each unit of risk. This can materially improve one’s portfolio results over time. In Part IV, which will follow shortly, we will cover a third portfolio approach beyond the two discussed here in Part III. In the meantime, if you are looking for additional ideas, WhoTrades Live offers a vast array of portfolios and strategies in how to manage your wealth. Each profile shows full transparency over virtually everything you need to know to make a decision as to whether it fits your personal money management needs and expectations. This includes returns, drawdowns, performance trends, holdings and percentage allocations, a computer- and/or user-generated synopsis of the trading style, popularity based on follower count, and any relevant trading activity.