DuPont – Question 1: Is The Return On R&D Investment Positive?

Print Friendly
Share on LinkedIn0Tweet about this on Twitter0Share on Facebook0


Graham Copley / Nick Lipinski



May 27th, 2015

DuPont – Question 1: Is The Return On R&D Investment Positive?

  • The key issue that DuPont’s board and management needs to address after its recent proxy victory is whether it is business as usual going forward or whether the recent events have caused a pause for thought. In our opinion, DuPont’s stock is not pricing in “business as usual”; it is pricing in changes – changes in the ways the company approaches both R&D spending and costs.
  • DuPont’s R&D is either horribly underproductive or it causes such a management distraction that the rest of the business has margins deteriorating much more quickly than for competitors. It cannot be neither of these options, as growth and returns can only be explained by one option or the other, or a combination of both.
  • In our view, DuPont’s belief in its “Science” strategy has caused the company to overestimate the value of R&D and of bolt on acquisitions and at the same time take its eye off the ball on costs more broadly. Competitors have eaten away at the cost curve over time, often using DD as the benchmark, while DD has essentially remained static.
  • The result is a deteriorating return on capital – as shown in Exhibit 1, made worse if we capitalize and depreciate R&D spend. Without the share price appreciation of 2013 and 2014, DuPont would only have returned 1% of its capital and cash spend to shareholders from 2000. With the share appreciation of the last 2 years it is 50% – still very value destructive.
  • A change of focus is likely needed at DuPont and while DuPont may have won the proxy fight, it needs to pay attention to what Trian was saying. We believe that the recent experience, coupled with fresh eyes on the board will result in more focus on what really makes money at DuPont and how to expand margins.

Exhibit 1

Source: Capital IQ, SSR Analysis


Some of the recent investor quotes following the DuPont shareholder vote highlighted some considerable shareholders support for DuPont’s R&D program. DuPont has a history of innovation and many of today’s household polymer and fiber products came out of DuPont’s labs. Today, there are many fewer new household names than there were in the 70s and 80s, with most of the best known now quite old; Corian, Kevlar, Tyvek, Nomex, Teflon; and others now in the hands of different owners; Nylon, polyester and Lycra, for example (Teflon will soon be with another company; Chemours). The newer products tend to be in the agriculture space with less catchy names.

For the last 20+ years investors have been asking DuPont whether the money spent on R&D drives a positive return, and while the company talks around the issue in positive tones, it is hard to tell. Over the last 20 years (1995 to 2014), DuPont has spent roughly $35 billion dollars on R&D, slightly more than its 2014 annual revenues. Since 2000, the company has spent $24 billion. If we think about R&D expenditures like capital expenditures and depreciate the capitalized R&D assets over 15 years, the current net R&D would be $13.6bn. If we assume that DD is getting a 10% after tax return on R&D, we end up with EBIT of $1.95bn – roughly the cost of the R&D program. If you subtract this from LTM EBIT of $4.4bn, you get a very low return on net capital excluding R&D. If you assume that all the R&D return is doing is offsetting the on-going R&D cost it is generating no net value.

If DuPont is making an adequate return on capital employed, then it is failing to do so on R&D or vice versa. In Exhibit 2, we show our TSR chart which shows cumulative capital outflows and shareholder returns from 2000 and in Exhibit 3 we sum the data for the 15 year period. It shows that DuPont has returned roughly 50% of what it has spent over the last 15 years to shareholders. Put another way, DuPont appears to have wasted half of the money it has spent. If we ignore the stock appreciation in 2013 and 2014, DuPont would have effectively wasted all of the money spent since 2000. Trian would argue that much of the stock appreciation over the last 18 months is because of its involvement and because of an expectation that there will be change.

Exhibit 2

Source: Capital IQ, SSR Analysis

Exhibit 3

Source: Capital IQ, SSR Analysis

DuPont has sold more businesses than it has bought since 2000, hence the negative net M&A bar on the chart, but has sold low multiple business and bought high multiple business, and has consequently lost earnings – the high multiple businesses have in aggregate not delivered the growth expected yet.

We would argue that all of this wasted cash/capital is a function of the R&D strategy! While the simple math clearly does not support the statement empirically, we would offer the following reasoning:

  1. While certain select R&D programs may have generated adequate returns, it is clear that the program as a whole does not and, in the aggregate, it destroys value
  2. Belief (optimistic and misplaced in our view) in the company’s overall ability to deliver returns and add value from R&D causes other capital allocation decisions that have not delivered value, or are taking longer to deliver value than expected – such as the Danisco acquisition. It is also likely that the multiple has been constrained as the company has grown more complex – a problem that we have discussed in the past.
  3. Management’s focus on the growth engine distracts from properly running the other businesses. Margins have fallen as more focused competitors eat away at the cost curve – this is Question 2 and will be a subject of further research.

This value destructive strategy has been questioned by us and many in the investment community for years and was the focal point of the message conveyed by Trian. Our expectation is that, post the proxy battle, and with new blood on the board, things will change. We believe that a more focused approach to R&D will emerge as well as more aggressive approach to costs. We have also
written about the latter several times in the past
and recognize that there is upside or downside depending on whether or not we are right. If we are wrong we will get more of the same. If DuPont sees revisions going forward that look like they did for 2015 – Exhibit 4 – the stock will lose momentum quickly and could easily lose 15-20% of its value.

Exhibit 4

Source: Capital IQ, SSR Analysis

We believe that investors should continue to ask questions about R&D productivity at DuPont to ensure that the company focuses its investments where it can make an adequate return.

Where Does All The Money Go

In 2014 DuPont spent $2.067 billion on R&D, around $100 million less than the prior year, and R&D expense has been around the $2.0 billion mark for the last four year. Over 60% of the spending is in the Agriculture business – see Exhibit 5.

Exhibit 5

Source: Company Reports, SSR Analysis

We published
a review of R&D productivity and value in the Industrials and Materials space
a couple of years ago, work which we will update shortly, and one of the metrics we looked at was cost per patent – the R&D budget divided by the number of patents issued. Exhibit 6 shows DuPont patents issued by year according to corporate reports. The company mentioned 600 patents issued in Q1 2015 in a recent conference call.

Exhibit 6

Source: Company Reports, SSR Analysis

In Exhibit 7, shown on the next page, we attempt to correlate “patent efficiency” with return on capital, the thought being that the more efficient you are with your R&D (cost per patent used as a proxy), the better return you might see. For the larger R&D spenders in the chemical space there is an apparent correlation, but also some outliers. PX, EMN and MON look efficient – others, including DD, less so. For this exhibit, we are using data from a central patent data base which does not align exactly with DuPont’s published data (though this may be a timing issue), but is consistent for all companies listed. If we used DuPont’s published data on patents, DuPont’s cost per patent would be roughly $1.5 million, but the ROC trend would not change

Exhibit 7

Source: Acclaim IP, Capital IQ, SSR Analysis

©2015, SSR LLC, 1055 Washington Blvd, Stamford, CT 06901. All rights reserved. The information contained in this report has been obtained from sources believed to be reliable, and its accuracy and completeness is not guaranteed. No representation or warranty, express or implied, is made as to the fairness, accuracy, completeness or correctness of the information and opinions contained herein.  The views and other information provided are subject to change without notice.  This report is issued without regard to the specific investment objectives, financial situation or particular needs of any specific recipient and is not construed as a solicitation or an offer to buy or sell any securities or related financial instruments. Past performance is not necessarily a guide to future results.

Print Friendly