Emissions trading scheme: The NZIER model

NZIER has released a very interesting report on the ETS (emissions trading scheme) in New Zealand (see their reports page – April). In this report they find that the economic cost of the emissions trading scheme will be eight times the cost of the “government paying” for the scheme. Although this eases to 4-5 times the cost in the long-run (which is the cost we should be interested in if we believe climate change policy will continue indefinitely) this is still a significant cost.

Although I think this is an interesting and important point to raise, and that it will add greatly to the debate to climate change policies – I get the feeling the result is slightly exaggerated by some of the underlying assumptions. Here’s the areas that I think are important to revist:

1. The cost of the “government pays scheme will be higher as a result of the income tax impact on labour supply,
2. The impact on dairy and the change in total emissions don’t seem consistent – implying that the cost of the ETS could be lower,
3. This number does not include mitigation technology – which is a steep assumption

NZIER do cover the third point in their full report, stating that with mitigation technology the cost of the ETS falls by about 8%. However, I will try to discuss the other factors below.

Now I will divide up the following long winded discussion into sections that can be read separately. If you have any questions or comments please tell me which section you are talking about.

Cost of Government pays scheme

There are two things that make me feel unsettled when looking at the impact of the government pays scheme.

Firstly, in the projections to 2025, the Kyoto liability is $2.2bn but the fall in GDP $1.4bn. I have been told this is related to the lower exchange rate track in the NZ pays vs doing nothing example – so I will just accept this. I would like it if someone could explain it to me though 🙂 .

This difference could also be explained by a change in government spending, making the track of government consumption differentin the two different cases – this could be a problem. The level of government spending should be (mostly) exogenous – as the level of government spending is determined to achieve efficiency and equity goals separately from the scheme. (This acts as justification for viewing the “NZ pays” scheme as a way of paying for it through additional income taxes).

Secondly (but actually highly related to that first assumption), they assume that the direct impact on the productive side of the economy from this scheme is zero – a very steep assumption. It is said directly on page 26:

This is because the Government’s offshore payment reduces domestic demand but does not impact directly on the productive capacity of the economy.

This is something I strongly disagree with. Off the top of my head I can think of three mechanism with which a increase in taxes (as this scheme effectively assumes a higher equilibrium rate of income tax) can impact on the productive side of the economy:

1. By influencing labour supply – as secondary earners are becoming increasingly important in the New Zealand economy, this impact matters.
2. By influencing savings rates and thereby investment – higher income taxes discourage savings (relatively) by decreasing the interest rate on savings. As savings=investment in the long run this will lead to a smaller capital stock.
3. By increasing the long-run cost of firm inputs (especially in industries where capital and labour are complementary).

As a result, I am certain that the cost of the “government pays scheme” will be greater than this report suggest.

Dairy, ETS, NZIER

In the long-run, NZIER assumes that their is a long-run yield associated with land and capital – a perfectly good and justifiable assumption. Given this they work out what happens to specific industries under the different arrangements.

The impact on dairy is substantial – the contribution of dairy farming to GDP is 13% while the contribution of dairy manufacturing to GDP is 12%. Without access to the CGE model or the assumption that went into it, it is difficult to question these numbers – however this would be one area that warrants further investigation.

A few of the issues I can point to with these dairy figures are:

1. It assumes we are a price taker in the dairy industry – however we account for a fair chunk of exports. As exporters are an important determinant of prices at the margin in many countries, New Zealand does have some market power for setting prices. This implies that if costs rise in NZ (which is what the ETS does), then the price itself will increase.
2. The responsiveness of dairy to the price increase is surprising. My belief was that much of our land was relatively good for dairy farming relative to other uses (especially given our low population). As a result, the long-run elasticity of supply in the dairy industry seems to be a bit more elastic than I think is fair. Again, if anyone can tell me the additional cost per unit of milksolids and put it in perspective insofar as the other costs of farming go we can have a clearer discussion of this.
3. However, given that carbon emissions don’t plummet, the quantity of diary production may be “relatively” unaffected by the scheme (I don’t know as the table is not divided into emissions reductions by industry) – this would imply that this result stems from a large increase the cost associated with production from the ETS. Currently, different interest groups say different things about the cost to dairy associated with the ETS – the final value we take will be important for working out the equilibrium impact.

How could a polluter pays scheme cost more than general taxation?

One of the interesting things to come out of the NZIER report is the fact that they say the polluter pays scheme will lead to greater forgone wealth than funding it through general taxation. This appears unusual when we view it through the lens of an externality tax (which is what an ETS scheme with no free allocation is like) where taxing those who create the problem improves efficiency while taxing those that don’t cause the problem creates inefficiencies.

Ultimately, this discrepancy can make perfect sense given our view on the increase in productive capacity. If we look at it in a static sense then an ETS would raise the funds at a lower cost to economic growth – however their explicit assumption that an ETS would hurt the capital stock but income tax increases wouldn’t, and the fact that many of our competitors won’t introduce an ETS on agriculture, changes the long-run natural rate of output between the cases.

NZIER’s illustration of how the ETS could be more costly is a good piece of work – solely for the fact that they illustrated a phenomenon that many people felt was counter-intuitive. Although their figures may exaggerate the discrepancies between the schemes the work is extremely useful nonetheless.

However, even if it is the case that the ETS is a more expensive mechanism than a flat income tax I’m not sure it is the inferior policy. In equity terms I believe that it is fairer for people who create the externality to pay for it then it is to get everyone else to pay. I think this is a point we should be careful to remember when debating what is the “least cost method” of carbon reductions – fairness has social value too.