The meaning of this example is that for negative P/Es, a higher absolute value (e.g., P/E of -100 versus -10) conveys smaller losses when the share price remains constant—creating a value bias of sorts. A recent ETF Market Intel piece on ETF.com highlights this point of view.

Flipping the example produces a different intuition, however.

But if we hold earnings per share negative, at say -20 cents, and vary the share price from $30 to $5, we get a linear inverse relationship, meaning negative P/Es with higher absolute value are priced more dearly for the same red-inked earnings—creating a growth bias.

What this means, unfortunately, is that relative to our benchmark, IBB’s P/E could suggest either smaller losses relative to price or higher prices relative to losses. These two viewpoints tug in opposite directions.

For this reason, data providers and issuers often avoid negative P/E ratios. Bloomberg, for example, displays “--” as IBB’s P/E ratio, effectively shorthand for negative.

IBB’s issuer BlackRock shows a positive P/E of 40.9. Blackrock omits the negative earnings from P/E calculations, most likely to avoid the challenge of comparing negative P/Es with each other, as I outlined the complexity above.

BlackRock also caps P/Es at 60 to squelch the soaring P/E values when earnings approach zero. As the chart above shows, tiny earnings can yield huge P/Es. Capping the higher positive P/Es limits the impact of these outliers—a reasonable step, in my view.

A quick peak at the earnings for IBB’s top 10 stocks showed negative earnings for BioMarin Pharma and P/Es above 60 for another four firms. These numbers flow through to our aggregate P/E number of -208.8 for IBB and explain why it differs so greatly from the issuer’s stated figure of 40.9.

So who’s right?

In my view, there’s no foolproof way to handle negative P/Es for ETFs.

When negative P/Es do show, they throw up a flag to investors, but beyond that, they’re hard to compare with each other.

Conversely, a method that uses reasonable rules and discloses them clearly—as BlackRock does—makes the numbers comparable, but could trip up casual or hurried investors by strongly implying a too-rosy earnings picture.

Bottom line: Investors need to understand what goes into their P/E number—as simple and familiar as it sounds—for the different sources they use.

*At the time this article was written, the author held no positions in the security mentioned. Contact Paul Britt at [email protected] or follow him on Twitter @PaulBritt_ETF.*