Good R&D, bad R&D: what's the difference?

By Ron Freedman

If there is a recurring theme in science policy legend and lore it is that Canadian companies don't spend enough money on R&D. International comparisons consistently put our business sector in the middle of the pack or lower in terms of its spending on R&D as a percentage of GDP (BERD). In reality, many sectors of the economy and many parts of the country meet or out-perform the OECD average in R&D spending, but the overall result is deemed to be sub-par. In the policy community, more (R&D spending) is always better.

In October, Research Infosource released its annual Canada's Top 100 Corporate R&D Spenders list. This year's ranking indicates that the country's largest R&D spender was Bombardier Inc., which devoted $2.2 billion to R&D in Fiscal 2013, up 15.4% from the previous year.

For most policy analysts Bombardier's increased R&D spending is good news. After all, the year-over-year increase added almost $300 million to the national business R&D tally and correspondingly expanded our closely monitored BERD/GDP ratio. So far, so good.

Bombardier a cautionary example

But was the increased spending good news for Bombardier? Herein lies a cautionary tale about innovation statistics. In Bombardier's case, a not-insubstantial portion of its Fiscal 2013 R&D spending was actually "bad spending" — un-planned and un-wanted R&D that was necessary to complete the development of its flagship C-Series airliner, after the program ran into difficulty.

If Bombardier's original plans had worked out the aircraft would now be in production and the company would not have had to devote scarce company funds (profits) to re-work its design and engineering. Talk to any Bombardier executive (or shareholder) and they'll tell you they wish they didn't need to spend the money. And yet, from a traditional policy perspective, Bombardier's bad spending on research is seen to be a good thing.

The Council of Canadian Academies' (CCA) recent report (The State of Industrial R&D in Canada, 2013) recognized this point when its authors observed:

"It is tempting to think of investments in (industrial R&D) as having a guaranteed rate of return, both at the firm and national levels. At the firm level, however, managers are well aware that R&D expenditures are a means to an end. In and of themselves, they represent a cost centre rather than a measure of success, and that outcome is never guaranteed." (p. 15).

In other words, given two identical companies with identical products, revenues, non-R&D costs, etc., the company that spends less on research to achieve the same objective will be more profitable and more successful in the long run.

In a sense, the argument is analogous to how we view GDP data. Regardless of whether increased GDP results from productivity and output gains in the real economy, or from an expanded black economy (e.g. gambling, prostitution, illegal drug sales), economists deem more to be better. Is it?

Companies invest in R&D for a number of reasons. The usual explanation is that they seek to develop new products or services and thereby expand sales and markets. Sometimes this R&D spending is proactive (gaining a step on competitors) and sometimes it is defensive (defending markets against new competition).

Many firms, especially in process industries, are looking for ways to reduce production costs and thereby enhance the bottom line. Sometimes, firms must respond (unwillingly) to new environmental regulations. This is simply to say that, from the firm's perspective, not all R&D spending is desirable. And by extension, not all R&D spending in the economy is desirable.

Given all the talk about the sad state of corporate R&D in Canada it's easy to forget that "Internal corporate funds cover(ed) $13.8 billion (or 85%) of all industrial R&D spending in 2012" (StatCan Industrial Research and Development: Intentions 2014). In other words, government spending accounts for about 15% of the total. Most decision-making on R&D occurs at the firm level, even if it is heavily influenced by government policy (e.g. SRED tax policy) or programs (e.g. IRAP).

I was asked in a recent media interview "What's the right figure for R&D spending?" which is what gave rise to this editorial. All things considered, the right figure is the lowest amount that a company needs to spend in order to remain competitive. Obviously, this will vary depending on the company, competitive position, industry sector, and a host of other micro and macro factors — the answer is very company-specific. The aforementioned CCA panel reviewing Canada's industrial R&D performance observed that:

Apple ranked 81st in total R&D spending [in 2010], and its R&D intensity (R&D expenditures relative to revenue) was only 3.1 per cent. In comparison, both Google and Microsoft invested over 10 per cent of revenues in R&D. Of the top 10 most innovative firms, only 3 were also in the list of top 10 R&D spenders … (p. 15).

Blackberry's high R&D didn't halt company slide

As a point of comparison, Blackberry Ltd, which finished in second place on our 2014 Top 100 Corporate R&D Spenders list, spent 18.9% of revenues on research. As a Blackberry shareholder, would that give you comfort? And yet for many in the policy community, out-sized R&D spending at Blackberry or elsewhere would be considered to be a good thing.

We can all agree that an appropriate level of R&D spending is to be desired – both at the firm level and at the national level. What's not desirable is a call for an across-the-board increase in spending regardless of need or appropriateness. In the final analysis, shareholders – and governments – need to have faith that business owners and managers will devote the right amount of available funds to R&D, taking their individual circumstances into account. The bottom line: More is not always better.

By Ron Freedman is CEO of Research Infosource Inc.