If you click on the Investment Philosophy tab on our website you will find this summary description:

“…the cornerstone of our investment philosophy is build for each client a well-diversified, risk-controlled portfolio that is appropriate for their specific situation and objectives. And to do that, we strive to build relationships with our clients to make sure that we both understand what the other is trying to accomplish.”

There are a lot of important elements to that statement with the most critical being building relationships with our clients. That said, in this article, I would like to focus on the phrase “risk-controlled.” Specifically I’m going to discuss what we mean by risk, how we measure risk, and how that is reflected in your portfolio.

Essentially Risk Means Uncertainty
One of the great questions in investment management is “what is risk?” There is, however, no one right answer. Investment risk may mean something different to each client. To some, it may mean the maximum loss their portfolio might suffer in a given time period. Others may define it as not keeping up with a particular benchmark or simply be failing to increase the real (after-inflation) worth of their portfolio. Still others may view it as the chance they won’t meet their financial goals.

And there are nearly as many ways to measure risk as there are ways to define it. Semi-standard deviation, beta, loss ratios, down market captures, and correlations are just some of the many statistical measures cited in the industry.

But whichever of these (or any other) “definitions” you subscribe to, what they all have in common is the notion of uncertainty. How much could I lose? Will I beat inflation? Can I achieve my goals? And indeed, uncertainty is the essence of risk (if I know for certain the outcome of an investment, there is no risk involved).

For us then, the broadest and simplest (and we like simple) definition of risk for a particular investment (or a whole portfolio), is simply the degree of uncertainty surrounding its future returns.

Uncertainty Is Best Expressed As Volatility
Just as no one definition of risk is perfect; no one statistical measure of risk is perfect, either. However, sticking with our broad definition of risk, the most popular measure of the uncertainty of returns is volatility, and that is our preferred choice as well. In statistics, an investment’s volatility is defined as the standard deviation of its returns. Without going into a lot of math, for simplicity’s sake, let’s just say that standard deviation is an indication of the “normal” range of possible returns. Under normal circumstances, 95% of the time the annual return of an investment will fall within a range of 2 standard deviations higher and lower from its average annual return. For example, an investment that has historically earned an average annual return of 8% with a standard deviation of 12%, could “normally” (95% of the time) be expected to deliver returns ranging anywhere from 32% (8 +12x2) to -16% (8-12x2) – a fairly high degree of volatility.

To make comparisons of volatility among different investments easier, we use a measure called relative volatility (see footnote #1). This is simply the standard deviation of the individual security or portfolio divided by the standard deviation of the broad stock market (S&P 500). A relative volatility of 1.25 therefore means that the investment has been 25% more volatile than the S&P 500, and conversely, a relative volatility of 0.75 means it has been 25% less volatile.

Why We Care About Volatility
Generally speaking, “success” in investment management is not about maximizing total returns, but about maximizing the investor’s chances of achieving their goals. These goals are usually defined in strict economic terms, (but not necessarily exclusively so), and reflect their unique constraints including temperament, risk tolerance, sophistication, time horizon, and current financial situation.

Numerous studies have shown that the ability to stick with their investment program is perhaps the most important variable in determining whether or not an individual will be successful in meeting their goals. And the destabilizing effect of volatility, both to the upside and the downside, is the leading reason why investors abandon their programs.

Every person is different. Some people have the temperament and personal circumstances to be comfortable with an aggressively positioned portfolio and some people don’t.

This is where an appropriate asset allocation strategy comes into play. At Kobren Insight Management, “appropriate” asset allocations are driven by time-tested relationships among stocks, bonds, cash and other asset classes. For each particular “mix” of these asset classes we have a general expectation of how volatile it should be relative to the market, as relative volatility relationships tend to be fairly stable over time.

How “Rel-Vols” Are Reflected In Your Portfolio
Regardless of whether we are pursuing an aggressive, conservative or balanced approach for you, generally, every client’s portfolio will feature the same basic themes in terms of market exposures. If we favor international stocks over domestic stocks, or prefer high-yield bonds to Treasuries, or even stocks over bonds, those preferences should be reflected in your account.

However, for a growth investor we will play those themes in a different manner than a conservative or balanced investor using a different targeted relative volatility as a benchmark. While all growth-oriented client portfolios should have similar relative volatilities, they will be higher than the relative volatilities for balanced or conservative investors.

The relationship among the relative volatilities for these different risk profiles should be fairly stable over time. If they are not, either we need to tweak our asset allocations, or more likely, one of our underlying managers is altering their portfolio. Either way, we need to control this risk, and we will adjust the portfolios accordingly.

Expectations
We believe that clients should have a well-informed view on how their portfolios should generally perform during various market environments. While we use relative volatilities to manage risk in a client’s portfolio, they can also serve as a quick way to get an idea of how your portfolio should perform.

For instance, if you have a balanced portfolio with a relative volatility of 0.50, then given it is taking half the risk of the market, a reasonable expectation is that its return should approximately be one half of the market or better. While there are some flaws to this back of the envelope technique, it is a valid way of looking at expected performance (see footnote #2).

Returning back to where we started with the statement of our investment philosophy, the importance of building strong client relationships should now be clear. Only with an excellent working relationship can we “make sure that we both understand what the other is trying to accomplish.” And with that understanding, we can create a portfolio for you with the proper relative volatility to make achieving your goals a greater 'certainty.'”

 

Footnote #1
Definition of Relative Volatility: A statistical measure of the approximate volatility or risk. It is the standard deviation of a portfolio’s daily total returns divided by the standard deviation of the S&P 500 stock index over the same time period. The S&P 500 has, by definition, a relative volatility of exactly 1.00. Relative volatility figures greater than 1.00 indicate portfolios which have proven riskier (more volatile) to hold than the S&P 500 index.

Typically, the industry norm is to use monthly data points to determine relative volatility. At Kobren Insight Management, however, we prefer to use daily data points. Not only do we think we can detect changes in portfolio market exposures more quickly this way, but we can also more quickly detect changes in risk characteristics, including relative volatilities. Using daily data also provides more data to work with, which increases our confidence in our statistical output.

There is, however, an interesting twist to using daily data over monthly data. The relative volatilities using daily data tend to be 0.10 lower (give or take a few basis points) than the relative volatilities using monthly data. We have not yet been fully satisfied by any explanation of why this consistently occurs.

Footnote #2
Technically, using relative volatility to determine an expected return is flawed. The industry measure that attempts to do this is actually “beta”. While beta does in fact incorporate relative volatility into its calculation, it also uses two other variables: correlation and the risk-free rate.

Nonetheless, given that Kobren Insight Management portfolios generally have high correlations to the stock market, coupled with the fact that risk-free rates have generally been low in recent years, using relative volatility to determine a crude expected return remains a respectable first cut at trying to understand portfolio performance.