Free exchange
IT IS a font of economic growth. Research and development (R&D), the hunt for new ideas, products and processes, enhances productivity and generates well-paid jobs. It plays an important role in keeping economies growing after they have industrialised. It has positive spillover effects on the number of employees firms hire, and on the wages of low-skilled workers in high-tech sectors. These public benefits help explain why R&D activity has historically been supported by large amounts of government cash. But public spending on research is in secular decline, leaving private firms to pick up the slack. This funding switch has already left Britain lagging in the global R&D race, and may yet put America at risk, too.
Private firms have plenty of incentive to shell out, of course. The companies that spend most on R&D tend to be in computing, cars and drugs. Top of their respective industries in 2010, according to Booz, a consultancy, were Microsoft, a software firm; Toyota, a car manufacturer; and Roche, a pharmaceuticals firm. Last year Microsoft spent $9 billion on R&D. Its 850 PhD scientists and 40,000 developers are mainly preoccupied by cloud computing. Toyota spent ¥730 billion ($9.2 billion); its hybrid cars trace their origins to 15 research centres which investigate new materials and fuels. Roche, which has over 330,000 patients enrolled in clinical trials, spent SFr8 billion ($9 billion).
Nonetheless, ensuring that private firms do enough R&D is a problem because its public benefits are likely to be higher than the private benefits (ie, profits) that go to firms. Even with patent systems to protect innovators, genuinely new inventions diffuse so that copycats and competitors benefit. These knowledge spillovers are good for the economy as a whole, but may be bad for the firm that made the original discovery. In a 2010 paper* that looked at data from American firms, Nicholas Bloom of Stanford University and Mark Schankerman and John Van Reenen of the London School of Economics found that the public benefits of R&D were roughly double the private benefits. The data also suggested that the optimal rate of R&D would be twice as high as the actual rate, and possibly higher.
Patent leathering
The simplest response to this problem is direct government funding. This was the favoured post-war model, especially in America during the space race. In the mid-1960s space exploration received around $25 billion a year from the taxpayer, measured in today's dollars. Public funding for military research was also buoyant. At its 1964 peak federal R&D hit close to 2% of American GDP. But the share of publicly funded research has since fallen sharply in some countries. In America the ratio of public R&D to GDP has fallen by half since the 1960s. In Britain the ratio fell by the same amount over a shorter period, dropping from 0.31% in 1986 to just 0.17% in 2009, the lowest level in the G7 group of advanced economies. This contrasts with sharp growth in public R&D spending in Japan, Germany and South Korea.
A fall in government spending places the onus squarely back on firms' private R&D efforts. In America this public-to-private switch has been relatively successful. Firms have increased their R&D spending faster than GDP, so that they have offset the reduction in government support (see left-hand chart).
Even so America's overall R&D-to-GDP ratio has fallen behind
South Korea's; China is catching up fast (see right-hand chart). Europe as a whole has a weak R&D-to-GDP ratio, despite German efforts. It currently stands at around 1.9%, far below a European Union target of 3%. Britain's position is particularly poor. Unlike in America, business R&D has fallen in tandem with public spending, dropping from 1.5% of GDP in 1986 to 1.1% in 2009. Of rich countries, only Italy looks worse.
Three things in particular help influence firms' R&D spending. The first is tax treatment. America was quick to offer tax breaks, introducing "research and experimentation" tax credits in 1981. This reduced firms' tax liability by 20% of their R&D costs, so $100m spent on research in effect cost only $80m. In a 2002 paper Mr Bloom, Mr Van Reenen and Rachel Griffith of Manchester University looked at R&D tax credits in nine advanced economies between 1979 and 1997, finding that a 10% fall in R&D costs produces a long-run 10% increase in R&D spending.
Merger policy is the second thing that can influence R&D. Competing firms race to be the first to invent new products; one of the rationales for merging with a rival can be to dull this competition. The pharmaceuticals industry has seen massive consolidation, for example: of 42 American drugs firms that existed in 1988, only 11 remained in 2012. Case studies suggest this may reduce R&D growth rates. In the six years before their 1999 merger, Astra and Zeneca increased R&D by an average of 19% a year. In the six years after AstraZeneca's R&D growth was just 1% a year.
Third, firms may face pressure from shareholders to rein in R&D budgets. Research is a current cost that delivers benefits in the distant future (although the drugs industry shows that pay-offs can disappoint). There is survey evidence that bosses choose to scrimp on R&D in order to hit earnings targets or to pay larger dividends to investors.
Britain's listless performance in private R&D spending is explained by all three of these factors. Britain was a tax laggard, introducing credits only in 2000. Much of its private R&D spending is down to drugs and services industries, where mergers are rife. And a recent government-sponsored report on equity markets found a lack of long-term decision-making by investors. But America should not relax. It looks increasingly stingy in its tax treatment of R&D as other countries catch up; mergers and short-termism are a concern in America, too. Relying on private firms to power R&D should mean giving them every incentive to do so.
.png)