Posts by Keith Ng
Last ←Newer Page 1 2 3 4 5 Older→ First
-
Whoops: I'd love to take you up on that. Want to give me an email at keith[at]point.org.nz?
-
Still trying to figure my way around forests... who knew counting trees was so hard?
Have a look here
Ouuu, excellent. Exactly what I need for the thing I was researching before this distraction.
-
I/S & George:
Whoops: yes; its a clear sign that they're not seriously interested in costing it, just producing a stick for the Minister.
There is a -15% / $200/ton scenario as well.
George: what Keith said. They simply did a "worst case" scenario - combining a strong target, an (even more) ridiculous price, and no international trade. Strap that chicken!</quote>
...
Bad economic modelling, which asssumes allocations as a shock, does not try to model the costs and benefits of pre-2020 adaptation to medium and high price signals, and leaves out forestry and other sequestration entirely , wins out.
Hey-o, I'm really hesitant about blaming the wonks on this one (I guess that's my bias). The worst-case-scenario was fine. You select a range which you model, then you do a worst-case-scenario *outside* of that range. It does not make sense to either do a worst-case-scenario inside the modelling range, or to extensively model stuff in the worst-case-scenario range.
The problem is when some fuckwit takes your worst-case-scenario and pretends it's the mid-point.
Still trying to figure my way around forests... who knew counting trees was so hard?
-
Hi Matt,
First, you're making the claim that a model which does not calculate the cost of domestic emissions reduction (DER), which the authors explicitly say should not be interpreted as the cost of DER, should be considered as the cost of DER. I think that the burden of proof is heavily on you.
There are two aspects to the model that make it deeply inappropriate to treat its results as the cost of DER.
1) The impact on RGNDI is the total cost of the AAUs multiplied by a factor of 1.7. This represents the cost of effectively “importing” AAUs, through its impact on our balance of payment and exchange rate (in addition to the cost of the AAUs themselves). By definition, 100% of the AAU deficit has to be purchased from overseas suppliers, so every dollar of the AAU deficit incurs this “import” penalty.
While a DER programme would involve importing technology and equipment, it would also involve domestic labour, generic local goods, as well as the potential for locally R&D'd and/or manufactured climate change technologies (which would also have spin-offs). The proportion of DER costs going overseas will be far less than 100%, and so the impact on RGNDI will be much lower, too.
$1 spent on DER is going to have less of an impact on RGNDI than $1 spent on purchasing AAUs. Of course, it might be may be cheaper to purchase an AAU rather than get the same amount of DER. So which one works out to be better? We don't know, because we don't know the import component of DER, so we don't know its impact on RGNDI. Which means, in essence, we don't know how much DER costs. Which is why this model cannot be considered a cost model for DER.
2) The model assumes a static climate change policy framework (see section 4.1.1, page 3). The DER it calculates is purely firm-level responses to the change in the price of AAUs.
It's modelling the cost of achiving the 40% reduction through nothing *but* the use of the ETS. Surely, that's an unreasonable assumption, right? Do I need to defend this?
Also, as George noted, their elasticity models are not designed for the price levels that such a scenario would require.
3) As an aside, I think the $15b figure is just plain wrong. In run 11 (-40% AAU @ $200), the upper-estimate of the impact on RGNDI is actually $8.86b. That's the upper-estimate of the highest price scenario with the biggest AAU deficit. Did they get the $15b figure from somewhere else, or did they just multiply $3,000 by a guesstimate-generously-rounded-up population of 5m?
Would be interested if you know where the $15b figure came from.
-
Sorry Matt, you raise some very good points, and it deserves a much more substantive and holistic response. Ignore all my haphazard responses so far. I'll respond properly later this afternoon.
-
Matt:
Ahem. To clarify myself some more (and dealing with your first point), the domestic emissions reduction calculations in that report assumes no policy change. That is, it is how the market will respond at varying prices of carbon, given current policies.
ARGH... gotta run. Will restate this properly later.
-
To clarify - at any given price level assumption, the amount of domestic emissions reduction is (more or less) fixed.
-
Hey Matt, I'll take the easier one (number 2) first.
Check out tables 3 and 4 on page 7. Domestic emission reductions are affected by assumptions about price. In the Infometrics model, they are fixed for any given price level. In the NZIER model, they move very marginally at a given price level.
You're right that the 87.7Mt figure is wrong. The assumed emissions level depends on the price of carbon. However, my point is that this is an analysis about how much the emissions deficit would cost, not an analysis about how much the reduction of emissions would cost, and that still stands.
-
The assumption seems to be that a 40% reduction is equivalent in price to $200. Which isn't unreasonable, it just isn't noted as one. Every paper I ever read written by economists consists of assumptions, few of them explicitly noted.
I don't think it was an assumption that prices will go to $200 if there was a 40% reduction - it was just a high-end scenario. "Where would we be if price was at $200 *and* we were faced with a 40% reduction?"
Whoops: It doesn't test for price @ $200 for other reduction scenarios because it doesn't consider $200 to be likely. That's why it's the only scenario at $200.
-
Don't bend over just yet. I'm looking at the report and haven't yet reconciled it with the statements Smith has made.
No - Smith is still very much spanked, but Graeme is right that my original wording is misleading. It implies that "cut emissions by 40%" and "no cuts at all" are the only two options.
Russell - the distinction is too subtle and confusing to leave in there.
--
CLARIFICATION:
The original post said that "According to the analysis that Nick Smith has been waving around, if we do not cut our emissions by 40% by 2020, we will incur a cost of $60 per person per week."
This was untrue because if emissions was reduced by, say, 15%, the cost would be lower, yet it would not be "cutting our emissions by 40%". The statement would only be true if 40% and 0% reductions were the only options available.
This does not impact on the substantive argument - that the $15b is a cost of *not* reducing emissions - and it does not impact on the fact that Nick Smith's absolutely misinterpreted the NZIER/Infometrics report.