S&P 500 2,441.20 17.28
Gold$1,224.80 $5.30
Nasdaq 6,253.81 61.92
Crude Oil $60,490.00      $-1570.00
Following Momentum and Outperforming on the Downside
NWM Momentum Fund
Interview with: Tim Ayles

Author: Ticker Magazine
Last Update: Jun 12, 3:11 PM EDT
One can outperform the market not by beating it on the upside, but in doing so on the downside. To avoid being caught in long downward moves, the NWM Momentum Fund switches between asset classes depending on the risk environment. Portfolio manager Timothy Ayles relies on a back-tested model that signals the risk appetite of the marketplace, the momentum in a name, and the best investments for the specific environment.


ďWe aim to manage a fund that outperforms the market on the downside. Our biggest goal is to navigate out of long protracted down moves and to find a way to be either breakeven or make money.Ē
Overall, we aim to buy whatever is working. If energy, biotechnology, technology and pharmaceuticals are performing strongly, then weíll buy these sectors as opposed to buying the S&P 500 or the MSCI World Index. If the broad-based assets are performing better than individual sectors, we will only own the broad-based markets. Currently, we have 30 ETFs in our universe.

Q: Can you describe your model in more detail?

A: The model has been tested over multiple markets and timeframes. Typically, the model is always trying to get us out of stocks. That means that we own stocks only when they are performing really well. Again, we have a pool of investments that we rank against each other. For instance, if we have 20 risk-on ETFs, I will look at the top four based on our performance measures. These could be China, Italy, U.S. small cap and biotechnology, for example.

But the fact that they are the best performing risk-on investments is not enough to buy them. I put them through one more filter, which measures the performance of these ETFs against the strongest U.S. Treasury ETF. So, if the short duration U.S. Treasury is the best performing risk-off asset, I will compare the performance of the short duration U.S. Treasury ETF against the top four risk-on ETFs that I am considering.

The result may be that only three of the assets are actually performing better than the short duration U.S. Treasury. In that case I am only going to buy three of the risk-on ETFs and I will substitute the fourth with the strongest performing U.S. Treasury.

So, even if the model points to a risk-on mode, the portfolio may not be 100% in risk-on assets, if the strongest performing risk-on assets are not performing better than the strongest performing risk-off asset. We always make sure that the strongest performing ETFs are still performing better than the risk-off asset. If there is a chance that the market is going down, I would rather own the best performing assets in the bond market as well.

We put every investment through a dual filter as we aim to buy whatís working. We canít guarantee that it will continue to work, but we want to own whatís currently working relative to everything else.

We have an absolute return mindset. If stocks and bonds are dropping massively, then cash may be the strongest performing asset that will actually provide some return when both stocks and bonds are dropping.

Q: How often do you review the strategy and the portfolio?

A: Our prospectus allows us to change the portfolio bi-weekly or quarterly. The difficult aspect of an active strategy is the whipsaw action of the market. If the market starts turning down and we donít get out quickly enough, we are going to ride along that down move. If we get out too quickly because of a short downward move, we would miss opportunities. So, the key is figuring out how quick our trigger should be so that we donít get constantly whipsawed.

Weíve tested both the bi-weekly and the quarterly options. We found better performance and lower drawdown with the longer holding periods, but we found higher win rates with the shorter periods. So we created a volatility measurement to identify the long protracted down moves that really damage peopleís accounts.

Basically, we try to measure volatility to see when we need to move quickly and when we would be willing to wait out to avoid a longer protracted move. The most challenging aspect of our model is the risk of getting whipsawed out of the market or staying in the market when we should have been out. I believe that our volatility trigger really helps to know if we should hold on or we should exit.

Q: How do you construct your portfolio?

A: During a risk-on mode, we typically have 60% in equities and 40% in fixed income. We can never be 100% in equities. We also have limits on our sector and industry exposure. But within our 60% equity exposure, we could be entirely in small cap U.S. stocks, for example, or we could have 15% in biotechnology, 15% in oil and gas, 15% in China and 15% in the S&P 500. On the fixed income side, we own investments like high-yield bonds, convertible debt or emerging market bonds. During a risk-off mode, we can be a 100% in U.S. Treasuries.

Currently, we own four equity ETFs with allocation of 15% to each of them, and four fixed income ETFs with weight of 10% each. So, we have a total of eight ETFs and 60% of them are in equities and 40% are in fixed income.

We would never be more than 60% U.S. equities in a risk-on mode, but we can be up to 100% in fixed income in a risk-off mode.

Q: How far back have you tested your model?

A: Weíve back tested the model with the ETFs that we have back to 2003, because thatís how far back the data goes. We did the 2000 collapse synthetically, through our own recreated data, because we didnít have ETFs that went back far enough. For the testing of the 2008 crisis, we used the actual ETFs to see how the models would have done during these periods of time. In both cases, the models transitioned out of equities and 100% into U.S. Treasuries and other safe fixed income investments.

Q: Does the model give warning signals on impending market conditions?

A: The one thing the model cannot predict is a situation like the Black Monday in 1987, when the markets went down 20%. The market tops are rarely an event; they are usually a process. Our measurement shows that typically people first get out of small cap stocks but stay in the larger-cap names. They get out of the riskier names and move into the bigger names, because they are getting worried. A bottom is usually an event, not a process, so it is more difficult to predict, but there are definite warning signs near market tops.

Again, our goal is to outperform on the downside, not to beat the market on the upside. Over the last three years weíve captured about 90% of the upside, while our downside capture is negative 19%. We have actually made money during the downside movements in the S&P 500 and thatís always been our goal.

I built the model based on historic data, so I donít know what the future drawdown is going to be, but there is a good chance that what happened in the past will continue happening in the future. For us, itís not about making more than the market on the upside; itís about losing less than the market on the downside.

  1  2

  More: Mutual Funds Archive

Sources: Data collected by 123jump.com and Ticker.com from company press releases, filings and corporate websites. Market data: BATS Exchange. Inc