There are those financial advisors and money managers who have posted and touted fake returns. Bernie Madoff had positive returns quarter after quarter without fail, and we all know his name because of his infamous Ponzi scheme. But it seems like month after month, I hear about another financial adviser or money manager who claimed unreal returns.

Back in the early 1990s, an attorney referred a young guy whom the attorney called “a computer whiz”. The “WHIZ” as I referred to him had developed an algorithm that he claimed would beat the market. He had found a backer willing to invest $1 million in his method. However, since the WHIZ wasn’t a licensed as a stockbroker or financial advisor, his attorney recommended that he use me to open the account and initiate the trades. I agreed. The algorithm was set to follow trends, buying stocks when the trend turned up and selling when the trend turned down. It was long buying only, moving in and out of cash as the trends changed.

The WHIZ had back-tested the algorithm by running it against historical data and showing the rules-based program would beat the market by a significant percentage. I no longer remember how far back the testing had dated, but it was substantial, at least 20 years and maybe as far back as 1929. Consistently, the model proved that during good and bad times the returns were better than the market. I do not remember if the WHIZ used the S&P 500 or some other index as his benchmark.

The investor opened the account with $100,000, promising the remaining $900,000 when the algorithm had proven itself for six months. With that, the WHIZ launched the algorithm; it was my job to execute the purchases and sales to the exact specifications of the program. For three months, the WHIZ called me every few days with new trades to make. While I had discounted commissions, there were significant trading costs. As the quarter progressed, I began to wonder, with the number of trades being executed, if the profits from the trades would outweigh the difference between the short-term and long-term tax differential. The back-tested model certainly showed it would.

For me personally, if the algorithm really worked and clients poured millions of dollars into it, I would make a lot of money just executing the trades.

The first quarter ended, and it was time to evaluate. I don’t recall the actual numbers, but essentially the algorithm showed it had done quite a bit better than the market. However, the actual client account was several percentage points worse off than the market and considerably lower than the algorithm would have predicted.

What had gone wrong? We examined the costs, and the trading costs were what the algorithm had anticipated. When I examined the trading, the algorithm followed the trends, but was always late when issuing buys and sells. If that was the case, the algorithm would have done worse than the market and would have had similar returns as the client. The only way, in my opinion, that the algorithm was doing better than the market is the WHIZ was manipulating the numbers. In other words, he seemed to be falsifying the returns.

But I would never know. The WHIZ put our project on pause, claiming that he needed to refine the algorithm. The client left the money in the account for a few months and then eventually withdrew what remained. I never heard from the WHIZ himself again, but several people told me he was advertising that he had made money for the client, a completely false claim.

False claims are not uncommon. For a more recent example, in 2014 the SEC charged F-Squared with falsifying their results:

“F-Squared has admitted that it misled its clients over a number of years about the existence and success of its core strategy.

“The [SEC] charged F-Squared with touting a seven-year track record in its AlphaSector strategy that was not based on strategies connected to real money, as well as with using faulty calculations to inflate results. F-Squared reported $25 billion in ETF-managed-portfolio assets as of September, according to Morningstar Inc.”1

Both F-Squared and the WHIZ publicized their model or algorithm results, not actual client returns. AFC uses the composite of all equity growth client accounts. It is no fun when account values are down or trailing the market.

It is not fun or exciting to report to clients their accounts are doing worse than the market. There have been six years when AFC has had to do that. Fortunately, there have been 11 years when AFC did better than the market. What we have learned from these experiences is that AFC client accounts have recovered. Studies show once a bottom is reached, in general, the market recovers between 1/3 and ½ within three months and recovers to new highs in 9 to 12 months.2 While AFC knows that past performance is no guarantee of future results, AFC clients in the past have seen a similar trend of recovery.

There is pressure to perform. Sadly, there are those in this industry who will falsify the numbers to make their performance look better than it is. I wish I could say it was a rare occurrence, but it is not. Our commitment at AFC is to report the actual performance. We do know there will be times when we underperform our benchmark, and we will report those times with as much accuracy as we do the good times. Our goal is to outperform over the longer-term, 5 – 10 years, because we understand that downtimes do occur. However, our real results show that we have achieved our long-term goal, and we will continue to seek to do better than the S&P 500 for all of our clients.

 

A. Michael Adams
President & Principal

1. THUNNICUTT. “F-Squared Pays $35 Million to Settle Claims It Misled Investors.” InvestmentNews. InvestmentNews, May 7, 2014. https://www.investmentnews.com/f-squared-pays-35-million-to-settle-claims-it-misled-investors-60281.

2. Renshaw, Edward. “Is the Stock Market More Stable than It Used to Be?” Financial Analysts Journal 51, no. 6 (1995): 81–88. http://www.jstor.org/stable/4479887.