engineered portfolio

Did our engineered indexes peak your investment interest; but did they still seem a little too risky for what you are comfortable with?

In those indexes we pick US stocks on a monthly basis; and we’ve made an effort to pick stocks across uncorrelated sectors.  But at the end of the day, they all still carry US stock market risk. Portfolio management seeks to reduce risk by spreading your investments across many uncorrelated asset classes like international equities, bonds, and gold. So let’s do that with our engineered indexes to engineer a balanced index fund!

In this post we’ll add an allocation to international equities, bonds, and gold to our diversified US equity engineered indexes.  What we’ll be left with are two very well diversified funds that have exhibited about half the risk of the S&P 500 while matching, and even beating its returns.

Adding Uncorrelated Asset Classes

Our engineered indexes are picking stocks in the US equity market.  Lets add a few other asset classes that are uncorrelated to US stocks; and let’s weight our asset classes as follows.

• 50% – Engineered Indexes
• 10% – Emerging Markets (EEMV)
• 30% – US Aggregate Bonds (AGG)
• 10% – Gold Trust (IAU)

Engineered Indexes

We will use both our 10 stock and 20 stock engineered indexes.  Both indexes seek undervalued companies with good fundamentals and low volatility per sector in the mid cap and small cap value space.  The 20 stock index is across all sectors while the 10 stock index narrows the scope to only consumer staples, health care, consumer discretionary, and energy stocks.

Emerging Markets

International equities have been very poor over the last 10 years; but in the long run they can offer decent uncorrelated returns. This is especially true for emerging markets.  We’ll discuss and demonstrate emerging market’s dominance over international developed market equities in another post; but the takeaway is that we will only use emerging markets for our international equity exposure.  As we discussed and found in our engineered indexes, boring and low-volatility stocks, have outperformed their higher risk peers and have done so with less risk.  So we will use iShares low volatility emerging market ETF.  Note that in the data below I had to use their EEM fund because EEMV wasn’t launched until the end of 2011. Since inception EEMV has beat EEM with less risk, check out the chart. Its also overweighted to our favorite, boring-yet-stable sectors, check out the portfolio.

US Aggregate Bonds

The US bond market has exhibited much more efficiency than what we found with US equities.  What I mean is that a broad bond fund has done a very good job at combining all bond asset classes and weighting them effectively to offer the best risk adjusted returns of its underlying assets.  We’ll discuss bonds a bit more in a future post; but if we’re only holding 30% in bonds, all we need is this fund that tracks the total bond market.  It has 5,886 bonds currently.  If we had a larger allocation to bonds we may consider adding short term treasuries and emerging market bonds; but again we’ll discuss that later. For the goal of our balanced funds, AGG is a very nice and simple option.

Gold

We found in our research and testing that a 10% allocation to gold can dramatically reduce risk, while sometimes even increasing overall returns with rebalancing.  On paper gold looks like a terrible investment on its own.  Its very volatile and its returns are pretty low over time.  But when added to an investment of other asset classes we found gold to be very effective.  Gold has a negative correlation to US equities; its price goes up in times of financial stress because its bought up as a “safe haven” asset. When the market is doing well the price of gold tends to go down. But its not perfectly a negative correlation, it still moves on its own and over the long run it has provided some returns.  Be wary of gold miners though. We’ll discuss this in a future post; but gold miners hasn’t offered the same level of risk reduction that the price of gold has.  During financial stressing times, gold mining companies have fared as bad as the broad us equity market, sometimes they’ve even done worse. So we’ll use iShares’ IAU ETF that seeks to track the price of gold.

Returns over the last 10 years

So how have these asset classes done in the last 10 years? Below I plotted the total return of a $10,000 initial investment on January 1st 2007.  These returns include dividends.

Engineered Balanced Index Funds Returns

Let’s take those asset classes and add it to a portfolio of 50% our engineered indexes. We keep our asset allocation in line with monthly buying-to-rebalance.  There are no sales of asset classes, only companies within our engineered indexes when they leave the respective index.  This is done to minimize the tax implications; it also reduces the number of transactions.  Below the total returns of an initial $10,000 investment are shown compared to the S&P 500.

The asset allocation fund with our 20 stock index has beat the S&P 500 by a few thousand dollars but it’s done so with a much straighter/smoother line. The 10 stock index version has handedly won (1.5 times better) and its also had a smoother ride compared to the S&P 500.

Risk Adjusted Return

Below I’ve plotted the average monthly return versus the volatility of those monthly returns for each of our asset classes and then our two balanced engineered funds. Both of our funds have had a reduced amount of volatility; but still outperformed.

I prefer looking at downside risk and annualized return which I’ve done below.  This represents the true amount of risk an investor felt and the true total return they earned. In this plot we see how our 10 stock engineered balanced fund has offered about 5% more returns per year compared to the S&P 500; and its done so with much less inherent risk. The 20 stock version is even less risky; but its only barely beat the S&P 500.

Now let’s look at the trailing 12 month alpha and beta of our funds compared to the S&P 500. Remember beta is an attempt to measure risk relative to the overall stock market. A negative beta means that an investment moves in the opposite direction of the overall market. A beta of 1 means that it moves in unison with the market. And a low beta means that it is correlated to the market; but it moves less drastically.

Alpha looks to quantify the level of outperformance for the level of risk required. In a purely efficient market, there should be no such thing as alpha… clearly we’ve found an inefficient area of the market. Check out the trailing beta and alpha of our indexes.

These balanced funds have had a beta of close to 0.50 over time, meaning they have about half the risk of the overall stock market.  Yet!… they have beaten the stock market because they’ve consistently offered an alpha.

Hedging: Another way to reduce risk

My preferred method of reducing risk is adding uncorrelated asset classes as we’ve done and demonstrated in this post.  But some may opt for a hedging strategy that looks to… hedge your bets!  These work by finding stocks to short sell, or bet against. When done correctly a hedging strategy will leave you with a market neutral exposure, meaning you have no direct exposure to the stock market’s risk.  Your “only” risk is in the fluctuation of your ranking/picking factors.

All the research I’ve laid out in the last few posts has been thanks to the tools provided in Quantopian’s development environment. I feel indebted to this company so I wanted to develop a strategy inline with what they’re interested in. After reading what they look for in a hedging strategy, I took our engineered indexes and adding a hedging element. Specifically I looked for overvalued and volatile companies with poor fundamentals. Then I short sell these companies, we bet against them.

I developed a few strategies and entered it into their contest. If you’re curious with how they perform you’re welcome to check the leaderboard. The first contest these will be eligible for is number 26 and then all future ones. They start every month and run for 6 months at a time. These might be viable options for institutional investors but for your average investor, adding uncorrelated asset classes like I laid out in this post is a better method of reducing risk.

Where can I invest?

As I stated at the end of our post where we originally engineered these indexes, there are two ways to invest in what I’ve just laid out.  You can copy it or you can tell us you’re interested by subscribing to our blog.  We may be able to set up a free email subscription service to tell you the companies that come out of these filters.  You can then chose to do with this information as you like.  If interested, you could buy these companies/ETFs for free in a Robinhood account. Or you could chose to just research these companies as they come out to build your confidence in the approach before allocating capital.

But let us know your thoughts! And please subscribe to our blog if you’re interested in hearing more about some lessons we’ve learned in our investment research. Happy investing!

Disclosures