Research and Development Spending:

A COMPARISON BETWEEN NEW ZEALAND AND OTHER OECD COUNTRIES. Review NZ Science Review, Vol 50 (1) 1993, p.24-27.

Keywords: Growth & Innovation;

Every economy faces a problem as to how much it should spend on each commodity. In the typical market the decision is made by a multitude of individual purchases. No one person decides on how much bread should be made, but in a rough and ready way, the outcome seems broadly efficient and fair (if the income distribution is fair).

But suppose the commodity cannot be effectively delivered in the market? How then is the decision to be made. One strategy is to privatise the commodity, leaving the demand and supply decision to individual actions. But that does not always give a good answer. and as a result too little or too much is produced, or that which is produced can go to the wrong people.

Reluctantly, perhaps, then, the government has to get into task of creating an artificial demand and possibly supply, usually through some bureaucracy funded from the public purse. This does not solve the query, but transfers the decision to some bureaucratic mechanism.

This is about where we have got to in the science debate. There are still sounds off right who would hand it all over to the market, although it is instructive that despite so sounding for many years they still have not given an intelligible account of the position which relates to the reality of research. Ignoring them, but not forgetting there are effective market solutions in some areas, we are left with the decision of the public contribution to research in level and form.

At this point a fundamental change occurs in the type of the argument. Those who advocate privatisation do not have any vision of what the level of supply should be. Rather they see the establishing of a process through which the level is chosen. It is a perfectly sensible approach given the assumption of a feasible market solution. No one has to say what should be the amount of bread baked. Simply setting up a market process is sufficient.

However once the public mechanism is introduced. there is no such process. although the inept may well seek one, perhaps muttering about “outputs”, and “purchasers”. Instead. the decision has to be made in terms of the outcome. If bread was supplied by the state, then the state would want to directly assess whether it was baking enough. Of course it is rarely easy to decide what is the appropriate level – in the case of bread it might involve nutrition experts, who would certainly quarrel, while economists would have some uneasy doubts about their recommendations paying enough attention to the issue of the opportunity costs. When it comes to research and development, it is even less likely that an expert panel will be convincing. especially as the experts are likely to be beneficiaries of the outcome.

So we might make comparisons. Consider a scatter diagram for 23 OECD countries, relating their real per capita GDP to their spending on R&D as a proportion of GDP. The best fit line shows there is a tendency for rich countries to spend relatively more than poor OECD countries on R&D. Nevertheless. while New Zealand is markedly poorer than the OECD average (82 percent on this 1988 measure) its spending on R&D is not only below average, it is also below the regression line. New Zealand ought to be spending around 1.5 percent of its production on R&D. It is only spending about 0.9 percent. On an international comparison we ought to be spending 60 percent more than we do.

One of the major gaps is in our business spending on R&D. The same scatter diagram with business spending instead of total spending evidently suggests that our business R&D is a disaster. Instead of spending the 0.8 to 0.9 percent of GDP the international comparison suggests, New Zealand is spending about 0.3 percent. That is partly a measurement phenomenon, because there is less incentive for New Zealand firms to report their R&D spending since they are not entitled to tax breaks. However Australian business which gets a 150 percent write-off (New Zealand’s is only 100 percent) reports only 0.4 percent of GDP spending on business R&D.

The problem is not with government spending on R&D, for the scatter diagram of government R&D spending against GDP per capita shows New Zealand is close to (but still below) the best fit line. However we also need to look at the mix of spending.

The other factor is the industry mix. We would expect Switzerland with its strong pharmaceutical industry to be spending more on business R&D than New Zealand. A major contaminant in all international R&D studies is defence spending, where obviously New Zealand has less of an interest. These points are considered further below.

Government Spending on R&D by Activity
Percent of GDP

Defence 0.01 0.07
Agriculture 0.25 0.12
Industry 0.10 0.21
Energy 0.01 0.05
Exploration of Earth 0.10 0.07
Industrial Infrastructure 0.01 0.07
Environmental Protection 0.03 0.03
Health 0.04 0.04
Social 0.03 0.08
TOTAL 0.58 0.72

Table 1 is the best I can construct. although it suffers from the omission of university research spending. The comparison involves six OECD countries chosen by Frank Edwards of the Ministry of Research, Science, and Technology: Australia, Denmark, Finland, Ireland, Norway, and, Sweden, small affluent economies, with a bias to resource processing.

While New Zealand spends 0.9 percent of GDP on total R&D compared to a 1.7 percent average for the six reference countries, with Ireland at the bottom (0.8) and Australia (1.2) the next above New Zealand. What the tabulation suggests is that New Zealand government spending is weak on R&D on defence, on social research, and on industry, energy, and industrial infrastructure. It is strong on the agricultural sector, and exploration of the earth.

New Zealand has a larger than average agriculture sector, which might explain about half the difference there. The rest may reflect the tendency for the New Zealand government to provide agricultural R&D, whereas in other countries it may be more an industry responsibility. However there may be definitional differences. (Where is fisheries and forestry? Perhaps food processing manufacturing is in agriculture.)

This definitional effect may explain part of the smaller R&D spending on industry and associated activities. but it also may also reflect our lack of priority ( even faith) in manufacturing as a part of our economic future. The vision seems to be of a sector which is a pale imitation of foreign manufacturers, importing their technology (and perhaps their skilled workers, if the sector needs any). An even bigger worry is the lack of business R&D which is normally concentrated in manufacturing.

The international comparisons suggest that our private sector R&D performance leaves a lot to be desired, although we cannot rule out that by a strange quirk of fate we have an exceptionally low technology industry – in which case successful economic development is probably impossible (without being preceded by major emigration). We need to tease out to what extent this technically impoverished manufacturing is a failure of the private sector itself; to what extent it is a failure of government policies towards the private sector, and to what extent it is that poor government industrial R &D spending means there are not the spin-offs into the private sector.

The relative emphasis of New Zealand government R&D spending on exploration of the earth is a puzzle. It cannot be explained by our relative share of the earth, because both Australia and Norway (among the reference six) have lower population densities (more earth per person – or GDP), while the ratio for Sweden and Finland is similar .Again we have a case for further systematic investigation. In the interim I tentatively suggest that we have an example of the New Zealand establishment’s greater ease with earth than people.

In summary, what the comparisons thus far have suggested is that government R&D spending is not too far off track for the low income OECD country we are, but the product mix is a bit odd, and needs to be reviewed, especially the little we are spending on industry and social issues. I am tempted to recommend an additional government R&D spending of 0.1 percent of G D P, say $80 million, about a fifth of which to be spent on social issues and the remainder on industry including infrastructure and energy.

But the biggest worry is the poor R&D spending by business. Over the last seven years there has been a lot of fluffing around by government – especially its more market advisers. It is time to deliver. Since R&D involves long lead times, I would be inclined to do something urgent, if only to signal the government commitment. A 150 percent deductibility of R&D would bring us into line with Australia (and may improve our statistic merely by improving industry reporting). Because of the imputation regime for corporation tax the fiscal cost would not be great, but it would direct manufacturing towards the high technology where is our future, if we are to have one.

I favour government support for private R&D because of the evidence that a firm, even an industry, is unable to capture all the benefits from its R&D, so it will undersupply if left to itself. Tax deductibility as an interim measure, but there is some common sense in following the Australian practice. Even the finance sector, that bastion of free markets opposed to government intervention. is wont to pursue this argument. Recently the New Zealand Bankers Association called for a government response to the Australian government’s introduction of tax concessions for overseas banking units, fearful they would lose business to Sydney. The bankers are on weaker grounds, because there is not the same externality effects as for R&D. But one must welcome the new pragmatism towards the market which they are exhibiting, assured that they are not merely seeking their own self interest at the taxpayer’s expense, but are recognising that their past anti-interventionist stance was ideological and inefficient.

These musings arose from reading Frank Edwards’s Research and Development Spending, a valuable book In terms of its international comparisons, and hopefully the first of a number of careful MoRSI’ studies which will become increasingly more sector specific. Edwards’s study adds another intriguing insight when it traces the R&D performances over the 1980s. While other countries have lifted their spending – from 1.4 percent of GDP in 1981 to 1.7 in 1989 – New Zealand’s has declined – from 1.4 to 0.9. Especially interesting is his observation that countries poorer than New Zealand have been deliberately accelerating their R&D spending. Only two countries dropped their ratio over the period. The other was the UK from 2.4 to 2.25 percent – now that its oil boom of the 1980s is over we can expect it to be one of the poorest performing economies in the OECD.

The weakness of international comparisons is not the problems of definition and error, nor of the need for careful detailed analysis. More fundamental is the possibility that every one is out of step except us. The rest of the OECD unwisely increased spending on government R&D and subsidised private activities. And how they suffered for such a recklessness: economic growth through the 1980s, overall falling unemployment, with some prospects of growth in the 1990s. If only they had listened to our far-sighted advisers, who ignoring any evidence were able to pontificate on R&D strategy, cramming it into an ideologically committed market framework.

Go to top