The cost-benefit of Australian genebanks

We’ve now received a copy of the 2007 report to the Steering Committee of Australia’s National Genetic Resource Centre entitled “Benefit-cost analysis of the proposed National Genetic Resources Centre,” as trailed in an earlier post. That’s the one that was said to posit a return on investment of 119:1 for the Australian pastures genebank. It makes for interesting reading, and we’ll try to summarize the main points here.

First, some clarifications, though. It in fact does not posit a return on investment (ROI) of 119:1 for the Australian pastures genebank. Sorry we gave that impression earlier. That is the ROI for the whole Australian National Genetic Resources Centre (NGRC), of which the pastures collection is just a part. Let me back up a bit. The Primary Industries Ministerial Council of Australia agreed in 2006 that the 5 existing, separate state-run genebanks (managing something like 180,000 accessions in total) should be amalgamated into a single (but actually two-node), national facility.

This was seen as an important pre-requisite for meeting Australia’s obligations under the International Treaty of Plant Genetic Resources for Food and Agriculture. As part of that process, The Allen Consulting Group was asked to consider the situation of no agricultural plant germplasm being conserved in Australia, work out how much the NGRC would be worth to Australia if that counterfactual were not in fact true, and compare that to the projected cost of setting up and running the NGRC.

The cost side of things was not that difficult to work out. It was estimated that the 2-node (crops and pastures) NGRC would cost about A$ 590,000 to set up, and A$ 3.5 million a year to run thereafter (compared to A$ 3.705 million a year for the 5-genebank system). Over 30 years, discounted at 6% real (whatever that means), that’s an outlay of A$ 51.7 million.

To get to the value of the system, it was necessary to make some assumptions:

• Continued access to germplasm held in Australia would enable historic trends in the growth of farm productivity to continue.
• Holding no germplasm in Australia would mean growth of productivity due to plant genetic enhancement would continue, but at a slower rate. Productivity gains due to other things would stay the same.
• The share of productivity gains due to plant genetic enhancement varied from 5% (sheep farming) to 30% (field crops).
• The share of productivity gains from plant genetic enhancement due to access to germplasm collections in Australia was 30%.

Put all that into your spreadsheet and you get a total of A$ 5 billion net present value to the productivity of Australian agriculture over the next 30 years. Add another A$ 1.2 billion in assorted benefits not associated with increased productivity (things like water quality, public health etc.), and you get A$ 6.2 billion (that’s an annuity benefit of A$ 364 million), which divided by the 30-year cost gives you that 119:1 ratio.

How confident are The Allen Consulting Group in their results? Well, they identified two key uncertainties: the annual discount rate and the share of genetic gain attributable to access to germplasm in Australia. Tweak the 6% and 30% values of these two things, respectively, and you get a range of productivity benefits over 30 years of A$ 1.2-14.7 billion around the canonical A$ 5 billion.

But when you think about it what the study actually did is work out the return to Australian agriculture of those 180,000 accessions being conserved, somewhere. The material could be held in places other than Australia, after all, and still be available under the International Treaty. Ah, but the study also figured that the cost of servicing demand by Australian breeders would be 20% higher if the average 14,000 yearly seed samples they requested had to come from genebanks outside the country, rather than the NGRC (because of stuff like quarantine etc.). So there were efficiency gains, which were factored into the benefits side of the equation.

So there you have it. Put another way, each accession maintained in the Australian systems will cost the Australian taxpayer A$ 21 a year over the next 30 years, and return A$ 194 a year to the Australian economy. I think: I get confused, but I think that’s right. That sounds pretty good, though I guess one should compare it with other sorts of investments the Australian government could make. Anyway, it seems the powers that be thought it was a bet worth making. And it’s not only Australians that should be grateful for that.