Lumber Liquidators: Cancer Risks Overblown

| About: Lumber Liquidators (LL)

Summary

The cancer risk from formaldehyde in flooring appears overstated. The CDC calculations are likely a worst-case scenario.

The CDC report assumes 24/7 exposure to formaldehyde as well as using the 95% percentile emissions levels rather than the average emissions level.

As well, the large chamber tests used only some of the manufacturer lots and is more heavily weighted towards lots with high emissions results in the small chamber tests.

Even after correcting the ceiling height, the average estimated additional cancer risk appears to be under 2 cases per 100,000 exposures, similar to the low end from the original report.

Q1 2016 business results are likely to be affected by recent events as Lumber Liquidators continues to suffer brand damage.

Lumber Liquidators (NYSE:LL) is in play again for Whitney Tilson, who recently reopened a short position based on his belief (among other things) that Lumber Liquidators' business has been devastated by the CDC report about the cancer risks from formaldehyde emissions and the legal liabilities arising from the cancer risks.

I don't disagree with the conclusions about the business consequences. Lumber Liquidators' business already had -17.2% comps in Q4 2015, and any talk about cancer is likely to further damage Lumber Liquidators' brand even if it doesn't sell Chinese-made laminates anymore. However, after reading the CDC's original report (posted on Retraction Watch), I think the cancer risk has been overstated in the media. The damage from having cancer mentioned in the same news articles as Lumber Liquidators is likely to be worse than any legal consequences. Thus, I retain my prior opinion about potential settlement results.

What doesn't get mentioned in media reports is that the CDC's cancer risk calculations from the large chamber test results are essentially a worst case scenario. The large chamber tests used only some of the manufacturer lots used in the small chamber tests, and these lots were weighted more heavily to the lots that produced the highest emissions levels in the small chamber tests. The CDC also used 95th percentile emissions levels and also assumed that there was constant exposure to formaldehyde when calculating the cancer risk.

Using the 50th percentile results, all the manufacturer lots (instead of just a subset), and assuming that people actually leave their rooms/houses sometimes, results in a calculation that the additional cancer risk is estimated at under 2 cases per 100,000 exposures using the large chamber test results, even after correcting for the incorrect ceiling height.

Small Chamber And Large Chamber Tests

I agree that it may seem more appropriate to use the large chamber tests over the small chamber tests to calculate the risks of adverse health effects, since the large chamber tests are apparently closer to real world conditions. The emissions increase in the large chamber tests may be closer to 3.5x the smaller chamber tests though, rather than the 6x that Tilson suggests. The reason is that the large chamber tests only involved some of the manufacturer lots (5 out of 11) from the small chamber tests, including the three manufacturer lots with the highest formaldehyde readings from the small chamber tests. Using a like to like sample comparison indicates a multiple range of 2.7x to 8.2x, with the multiple tending to be less pronounced when the small chamber test already indicated a high emissions level.

Formaldehyde Emissions In μg/m2 -hr

Sample

Large Chamber Average

Small Chamber Average

Multiple

1

530

157

3.4

4

192

37

5.2

5

252

93

2.7

6

498

157

3.2

10

136

17

8.2

Average

321

92

3.5

Click to enlarge

Note that these are the original numbers and do not take into effect the impact of the incorrect ceiling height input.

Formaldehyde Decline Rates

There has also been some debate about whether formaldehyde emissions levels continue to decline over time or remain at elevated levels indefinitely. However, the small chamber test results in the CDC report would seem to favor the position that emissions levels decline over time. It tested samples that were manufactured in 2012, 2013 and 2014. There was "more variability and higher maximum concentrations in the 2014 samples than other years, less in 2013, and the lowest concentrations in 2012." The samples do involve different products and are not directly comparable, but the patterns do match up with the idea that formaldehyde emissions levels continue to decline for years.

Click to enlarge

Another long-term example that indicates that formaldehyde emissions levels do appear to continue declining over time in the real world comes from the FEMA trailers. A FEMA trailer that was tested in 2011 and featured in a story registered formaldehyde levels of 105.6 ppb, which went down to 20 ppb when it was tested again in 2015. The trailer would have likely been manufactured around 2005/2006, so it was already several years old when the first test was done. Formaldehyde emissions level decline rates do vary a bit according to the type of wood product, but it appears that the emissions level generally goes down by around 50% every 18 to 24 months.

Cancer Risk

The CDC calculates the cancer risk based on a constant exposure to the same level of formaldehyde emissions for two years. It also used 95th percentile levels for the formaldehyde emissions, which meant that 95% of the laminate flooring in question would be projected to result in formaldehyde concentrations at or below the number the CDC used for its cancer risk calculations.

Using the average emissions level instead would more than offset the impact of the original ceiling height calculation. I'm not entirely sure what the average emissions level is with the small chamber test, but somewhere around 15 μg/m3 seems reasonable for the average formaldehyde concentration result, given that the median is 8.92 μg/m3, and the average is going to be higher than the median due to the far end of the curve. Adjusting for the correct ceiling height would bring this to around 49 μg/m3, and then using a 3.5x multiple to convert to the large chamber results would give a number of 172 μg/m3.

Click to enlarge

Lumber Liquidators previously mentioned that its average laminate transaction involved slightly less than 500 square feet of flooring. This is only around 20% of the floor space of the typical home. It is unknown how much of the laminate ends up in rooms that people spend a lot of time in such as bedrooms, versus relatively little time such as the foyer. If we assume that 40% of inhabitants' time at home is spent in rooms with laminate flooring and assume that Americans spend around 69% of their week indoors at home, this results in the average purchaser spending around 28% of their time in a room with Chinese laminate flooring.

Using the estimated average large chamber test formaldehyde concentration of 172 μg/m3 (including the correct ceiling height) and the assumption that the average purchaser spends 28% of their time in a room with Chinese laminate flooring results in an estimated additional risk of under 2 cases per 100,000 exposures based on the CDC risk numbers for formaldehyde. This is at the low end of the original CDC report.

Conclusion

As the CDC is understandably biased towards health protection, its cancer risk estimates are a worst case scenario that models 24/7 exposure to formaldehyde and uses the highest emissions results as their default input. Recalculating based on real world activity patterns and using the average emissions results would result in an estimated additional risk that is relatively limited and certainly much less than some of the fear-inducing numbers being thrown around.

I continue to believe that challenging business results are the main threat to Lumber Liquidators and the recent events are certainly doing the company no favors. Even though Lumber Liquidators no longer sells Chinese-made laminates, past results indicate that brand damage affects other products. I am trading in and out of Lumber Liquidators depending on its share price, and do not anticipate good Q1 2016 results.

Disclosure: I/we have no positions in any stocks mentioned, but may initiate a long position in LL over the next 72 hours.

I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.