Image for Economic Effects of the Pulp Mill – A tale of two assessments

In last week or so, there have been two published assessments of the economic impact of Gunns pulp mill on the Tasmanian economy.

One is based on a silly mistake.

The other raises more questions than it answers. 

Sunday Tasmanian

The first can be disposed of fairly quickly. It appeared in an article by Dr Bruce Felmingham in the Sunday Tasmanian, 13 March.

According to the article, impact studies show that ‘$6 billion income will be generated in the form of gross state product annually …. [which] … represents 20 per cent of the current value of gross state product’.

Were this claim to be true, Ms Giddings could solve her budgetary problems immediately.

She should build the mill with government money. A $2.5 billion investment would yield an annual return to gross state product of $6 billion, with a nice little share flowing back to the budget via taxes. 

Budgetary problem solved!

The error in the Sunday Tasmanian flows from a misunderstanding of what the $6 billion figure actually represents.

It is not an annual flow. It is the sum of the discounted additions to gross state product over the modelling horizon (in the Gunns report to the RPDC, this was 2007 – 2030).

No doubt the Sunday Tasmanian will publish a correction in due course. 

Insight Economics

Now turn to the Insight Economics report released on Friday 18 March. 

This is more substantial. It is an updated version of the Gunns report to the RPDC, conducted using the same Monash University model as before, but with the modelling period advanced from 2007-2030 to 2011-2030.

As such, it is subject to the same strengths and limitations as the earlier report to the RPDC – its strength is that it’s based on a model structure that is reasonably well understood by practitioners.

Its limitations are that the underlying assumptions are unclear, it doesn’t allow for risks, it doesn’t allow for adverse environmental effects, and so on. Since I’ve made those arguments at length before, I won’t rehearse them here. 
There are similarities in the modelling assumptions:

• As far as can be ascertained the technical specifications of the mill are the same – in both cases the mill has a capacity of 1.1 million tonnes of air dried pulp. In the earlier report, it was assumed that the mill would be operated to produce between 820,000 and 960,000 tonnes of pulp per annum. The assumed production rate in the Insight Economics report is not known.

• As before, it is assumed that the debt finance for the company is sourced from the global capital market. So debt interest (and much of the dividend flow) will flow offshore.

• The cumulative monetary impacts are discounted at a rate of 5%

There are a few differences:

• Earlier, the wood supply was a mix from Forestry Tasmania and other Tasmanian sources. Now it is envisaged that it will be plantation based and that 10% of the wood supply will come from the mainland. But in both cases, the modelling assumption is that the mill will not, of itself, change the ‘base case’ Tasmanian wood production scenario. 

• The exchange rate has appreciated significantly since the last report – in 2007 the Australian dollar was trading at around 85 US cents; now it is near parity. The long-term exchange rate assumption used in the modelling is, in either case, not known. 

• The mill now requires fewer workers to operate – before it was 292; now it is 250.

The table below compares results from the two reports. Because the earlier report measured things in terms of 2005 prices, I’ve inflated the earlier numbers by 13% so as to make comparisons in terms of 2010 prices.


Although the exchange rate has appreciated and it is now proposed to import wood from interstate, the Insight Economics report paints a much rosier picture than before:

• it is calculated over a time horizon that is four years shorter, but the cumulative effect on gross state product is 30% larger (9.9 billion dollars compared to 7.57 billion dollars).

• cumulative investment is higher – but only by about the same amount as the increase in the assumed construction cost of the mill (1.45 billion dollars versus 2.3 billion dollars ).

• the impact on consumption is about the same.

• the Tasmanian balance of trade has increased by about 1.5 billion dollars.

• the indirect employment impact is more than double. Now the mill creates 3100 jobs (250 direct and 2850 indirect), compared to the earlier estimate of 1617 (295 direct and 1322 indirect).

One explanation that could account for the differences is that the mill is now assumed to operate at full capacity, giving a 15% increase in output of pulp. On the other hand, 10% of the wood supply is now going to be imported, cutting the monetary and employment impacts on the Tasmanian economy.

Alternatively, there might be a brighter outlook for world pulp prices. Given the appreciation of the Australian dollar, it would have to be a lot brighter to generate this sort of improvement in gross state product.

But in any case while higher pulp prices might lead to higher profits (and hence higher gross state product), most of the dividends will flow interstate or overseas. Higher profits would not translate into higher employment effects, purchases of raw materials from Tasmanian suppliers, or induce greater activity in the Tasmanian economy.

So the results of this report, which finds that the mill will produce more than twice as many indirect jobs as earlier estimated, are a puzzle.

This isn’t the first time that successive pulp mill modelling reports have produced wildly different results – see my earlier post: ‘Da Vinci, Picasso and Minister Tony Burke’: HERE

That won’t stop some parliamentarians and journalists reporting the latest set of results as holy writ. 

But to convince the rest of us it would help if, just this once, the proponents and their consultants were more forthcoming on their modelling assumptions.

First published: 2011-03-19 07:46 AM

Noel Pearson, The Australian: Backroom deals bless their wildest dreams