Building Queensland and Infrastructure Australia require the use of a 7% discount rate in cost-benefit analyses of public infrastructure projects, but is a 7% discount rate applied to the future benefits of public infrastructure projects too high? Marion Terrill of the Grattan Institute argues discount rates should not be frozen where they were in the 1980s, given that real interest rates paid on borrowings by governments have fallen substantially since then. I’m grateful Marion agreed to be interviewed for my podcast. Our conversation is now available as Economics Explained EP42 Unfreezing Discount Rates.
A 7% discount rate means that $100 of benefits in fifty years’ time are worth only $3 today, in present value terms, compared with $18 (i.e. 6X higher) for a 3.5% discount rate. A higher discount rate means we are less likely to favour projects with high upfront capital costs but which deliver benefits to current and future generations which grow over time. Arguably, a project like Cross River Rail falls into that category, although I’ve always been sceptical about the projected benefits.
Here is a link to Marion’s 2018 report on discount rates co-authored with Hugh Batrouney:
Unfreezing discount rates: transport infrastructure for tomorrow
The Resources for the Future Working Paper I quote from in my introductory remarks is Discounting for Public Cost–Benefit Analysis
Queensland Economy Watch 
Great point, thanks Gene.
A similar point on the discount rates applied to health prevention investment was made recently in this article…
https://www.publish.csiro.au/AH/pdf/AH20057
Mike
Thanks Mike. I’ll have a read of that CSIRO article
Hi Gene, yes, discount rates have been on my mind latey. Looking at an ASX company with gold assets in Africa (Ghana) and they a 5% discount rate to put an NPV for an PFS! I had to question that. I don’t think mining companies in Australia use a rate that low. More like 7 to 8% now. But for the government projects in this discussion then perhaps 7% is too high. We do have a financial benefits but also unquantifiable community benefits as well. The only issue is that we are looking very long term and no one knows the future. So, picking a discount rate now to very low interest rates may overlook the future risks. Eg the community in semi-permanet fear of virus transmissions will cut public transport use so freeways and road tunnels would be a better option.
Thanks Russell. Yes, we definitely need to consider future risks. Let me post on this again soon because Marion made an interesting point in our conversation I want to look at more closely.
Great topic Gene
Warning. Nerd comment ahead.
I don’t disagree from a pragmatic point of view that using a social discount rate of 7% is probably too high for pure public good projects, but the arguments always seem to be driven by the fact the cost of debt has fallen (i.e. a change in the cost of a financial instrument is being used to argue for a change in an economic parameter – the social discount rate).
Some readers might remember a mention of a genius called Frank Ramsey during their one lecture on social discount rates at uni. So what would the social discount rate be if we went back and reworked the ‘Ramsey rule’ (i.e. we used measured economic parameters and not financial instrument prices to drive the calculus)? One of my team did that for a project a couple of months ago using Australian data from the last 10 years and it suggested a social discount rate should be quite a bit lower than 7%, but not as low as the estimates developed by Resources for the Future (who used a more complex approach, different assumptions, and US data).
There seems to be a body of evidence that infers the 7% central discount rate suggested by Treasuries is possibly too high, and deserves some attention. I’d really like to see some proper analysis from some real experts on this topic. It is fundamental to so much of the work we all do.
But a word of cynical caution. It seems to me that much of the advocacy for using lower discount rates is from entities suggesting that governments should go into debt to finance marginal (at best) infrastructure projects. Are the cheerleaders of infrastructure projects looking for an alternative means to fudge the figures now that they have been called out on the practice of systemic ‘optimism bias’?
haha, that’s a great nerd comment, Jim! Marion and I discuss the optimism bias issue in our chat and acknowledge it’s a legitimate concern. According to Marion it’s the reason that’s been advanced by Infrastructure Australia and other bodies for not lowering the discount rate. But is it right to set too high a discount rate rather than more rigorously evaluating purported benefits?
Gene
I was taught that you never adjust the discount rate to make up for a lack of robustness or incomplete analysis of the benefits and costs. You are just replacing one flaw in the analysis with another flaw.
We should all be striving to get the identification, scoping and assessment of marginal benefits and costs estimated correctly in the first place.
I absolutely agree Jim. Thanks for the comment.