In science, it's easier to disprove a universal hypothesis. Youonly need to find one counter-example to bust the entirehypothesis. You might hypothesize, “Man-made industrial era carbonemissions are solely responsible for causing global warming.” Thatis a universal hypothesis because, by identifying a single warmingperiod that occurred prior to industrialization (e.g., the MedievalWarm Period), you can disprove this universal hypothesis.

You might say, “Why not add appropriate hedging language to makeit harder to disprove the hypothesis?” A non-universal hypothesismight be: “High carbon levels in the atmosphere may increase globalwarming.” This wording makes it more difficult to disprove thehypothesis. You can find a period where high carbon levels existduring an Ice Age era (some research suggest this may have been thecase), but that does not disprove the hypothesis (since thehypothesis contains the word “may”). In addition, the hypothesiscontains no baseline to reference any increase. Failing to see anincrease may not disprove the hypothesis because temperatures mayhave actually been lower if carbon content would have beenlower.

This represents a hypothesis that's virtually impossible todisprove. You might think this is an ideal hypothesis, but then youwill have exited the realm of science and entered the world ofpolitics. In science, the only purpose of creating a hypothesis isto discover a theory, i.e., a universal truth. To discover auniversal truth, you need a universal hypothesis. Any lesserhypothesis has no value in science or truth, telling you a lotabout politics.

Which brings us to mutual fund expense ratios.There's a popular meme which implies you're better off picking afund with a low expense ratio. This is generally followed by a memethat concludes, because index funds have low expense ratios, youshould only invest in index funds. If you convert either of thesememes into a universal hypothesis, it's incredibly easy to disprovethem. Let's do that.

In the first case, the universal hypothesis would maintain that,“Low expense ratio funds willalways perform better than higher expense ratio funds.” We onlyneed to find a single example of a higher expense ratio fundperforming better than a low expense ratio fund to thwart thishypothesis. In looking at the 15 year return data for theMorningstar database ending December 31, 2014, five out of the top10 performing funds had an expense ratio above the average. In all,25 funds with an above average expense ratio had above averagereturns. Hypothesis denied.

In the second case, we need only find a single case where anactive fund with a higher expense ratio active fund outperforms anindex fund with a low expense ratio. The sameMorningstar database shows 95 of these actively managed funds beatthe best performing (and lowest expense ratio) S&P 500 indexfund for that period. Hypothesis denied.

A salesman can sometimes get away with using disprovedhypotheses. A scientist can't. So, is a fiduciary more like asalesman or a scientist?