Matt McIntosh is too kind to Colby Cosh

"Africans aren't helpless animals--they know what works against malaria"

More accurately, they don't, but the Conservatives are here to tell them, by God!!

The choice between nets and spraying is, in any case, largely independent the choice of insecticide. You can treat nets with DDT or spray with other insecticides.

"Sumitomo Chemical (SCC) will be donating over 330,000 Olyset® anti-malaria bed nets - worth around $2 million - to the Millennium Villages in sub-Saharan Africa. One
hundred villages (individual population c. 5,000) in 10 African countries will receive the Olyset nets, enabling at least half a million people to be protected from exposure to malaria."
...
"SCC's Olyset is a long-lasting insecticidal net (LLIN) recommended by the World Health Organisation. It never needs chemical treatment by the user and is the only long-lasting bed net guaranteed to last for at least five years."
http://www.sumivector.com/images/pdf/060327-sumitomo-millenium.pdf

I don't think I was too kind. There are significant nuggets of truth to this stuff, as Greg Cochran pointed out in the comments to the post -- environmentalist lobbies have done real harm by stigmatizing DDT, discouraging donor countries and NGOs from funding perfectly good indoor use of it. I also think the odds of Cosh influencing someone's donation choices with that comment are close enough to an engineering zero that there's not much reason to be too harsh about it. Bullshit bugs me, but there's too damn much of it out there on this issue (much like global warming) to get angry at someone for getting it wrong.

From the Malaria Journal paper:
"Arguably the most cost effective tool in malaria prevention is the use of ITNs. Randomised studies have documented up to 30% reduction in the number of under-5
deaths through ITN use alone [20]. In a related report it was concluded that 6 deaths are averted for every 1,000 children age 1-59 months that sleep under ITN [4]. In Eritrea, ITNs use as a single intervention was strongly correlated to the 84% decline in morbidity and mortality. In view of the high effectiveness of this method three related issues needed to be assessed: usage of nets by children and pregnant women, re-treatment of ITNs, and
sustainability."

Or, as Mr. Cosh puts it,
"However, the program in question incorporated both bed nets and indoor DDT spraying, and the study doesn't address the question of the relative cost-effectiveness of interventions."

Apparently, Edmontonians, unlike Africans, are in fact incapable of grasping what works against malaria.

"It's Tim's decided opinion that I am full of shit on this issue."
Wonder why?

Apparentl

Matt:

the thing is, with many of those DDT fanatics, it's not that they get their facts wrong, it's that they deliberately get their facts wrong. DDT as malaria killer is a religion to them, not something to be questioned.

Why is this proof that nets are a good substitute for DDT? Eritrea IS using DDT.

To the larger question of whether or not DDT is effective, here's a graph showing low levels of malaria during the period of heavy spraying (block A) and high levels of malaria during low levels of spraying (block B).

http://www.cdc.gov/ncidod/EID/vol3no3/robert_8.gif

Hans, they used ITNs as the primary method. There wasn't a statistically significant correlation between DDT spraying and the reduction in malaria.

In some places DDT is effective, in some places it isn't. It depends on a lot of factors: what is the vector? what is the vector resistant to? is transmission unstable or stable? is there infrastructure to support a sprayin program? etc etc

It's interesting that when presented with arguments that other pesticides are as effective as DDT, the spreaders of the blood-libel will insist that the lower price means DDT should always be used.

When pesented with evidence that nets are more cost-effective than DDT spraying their reaction is "use both".

Martin, DDT is a malaria killer under many circumstances, when used properly. Even Tim acknowledges that. And I would not be too quick to cast the first stone. Of all the many people I've interacted with online, I can count on one hand the people who I've never caught talking dogmatically out of their ass on some subject at least once. That's myself included, and I bet you're not an exception either.

"There wasn't a statistically significant correlation between DDT spraying and the reduction in malaria."

I don't get that from what you've provided here. In fact, I get the opposite:

"Malaria morbidity was strongly correlated to the numbers of ITNs distributed (SZ=-0.125, p<0.005) and the amount (kg) of DDT and Malathion used for IRS (SZ =-2.352, p<0.05)."

Hans, that's a correlation with DDT+Malathion. There was a significant correlation with Malathion alone but not with DDT alone. With Malathion alone the correlation coefficient was higher than with DDT+Malathion, which suggests that Malathion was more effective than DDT. This might be the reason why they've been reducing DDT use and increasing Malathion use.

I am not getting that from the study you linked. Are you thinking of another study? This study suggests the opposite of your conclusion on the effectiveness of nets (see last paragraph) and says nothing about the relative efficacy of the two insecticides used:

"Studies on vector resistance to DDT and other commonly used insecticides revealed high efficacy throughout the period. No resistance was reported."

"There was no resistance detected to DDT, Malathion or any of the insecticides used for the control of
adult mosquitoes."

"IRS was the next most important vector control method in the country. Approximately 13% of the population in malaria risk areas of the country benefited from IRS. DDT and Malathion were the two chemicals commonly in use. Effective community mobilization and involvement contributed to the observed increasing IRS coverage. Controversies surrounding the use of DDT, which was the mainstay of eradication and vector control, have tended to undermine success in the tropics [1,23]. Recent shifts in favour of controlled indoor use of DDT have supported renewed interest leading to its reintroductionin some countries including Eritrea."

"Eritrea advocates for the use of DDT for IRS alongside experimental preparations. Although the use of DDT is contestable given the environmental risks it poses, many of
the poor countries cannot afford the alternative chemicals currently being tested, as their cost is prohibitive [24]. In addition tests in Eritrea on resistance to the commonly used insecticides do not show any evidence of resistance to DDT."

"The final question to be explored was the role of each of the interventional measures in reducing morbidity or mortality. Within the limitations of the current study design it is evident that combining ITN use with IRS or other vector control measures did not confer added value to the outcome in malaria mortality or morbidity. This is not surprising since An. arabiensis is endophilic, and both methods act at the point of breaking the vectorhuman
contact. This is supported by observations from elsewhere that DDT spray can eliminate up to 93-95% indoor resting density of a vulnerable vector [23]."

http://www.malariajournal.com/content/pdf/1475-2875-5-33.pdf

Lambert, you really should finish off the discussion at Gene Expression. As it stands, you're letting people get away with calling you a liar.

(Go to http://www.gnxp.com/blog/2006/04/ddt-myths.php and click on comments.)

Hans, see Table 8.

"When pesented with evidence that nets are more cost-effective than DDT spraying their reaction is "use both"."

And of course, the whole argument over the cost-effectiveness of DDT is just a side issue they bring up when you inquire just exactly how is it DDT is "banned" when their own arguments derive from its use?

"It's interesting that when presented with arguments that other pesticides are as effective as DDT, the spreaders of the blood-libel will insist that the lower price means DDT should always be used."

Particularly since the development of permanently treated bed-nets, no matter how cheap DDT is the costs involved in hauling drums of the stuff over dirt roads to to small villages for periodic respraying has got to count against it, cost-wise.

Tim,

Where are you getting these numbers? They are not in that table. You are taking positions unsupported by the authors, which never stated that nets are more effective than DDT (in fact they implied the opposite). Nor did they ever state that Malthion is more effective than DDT. Nor does the data itself support these assertions.

The p-values of DDT + Malthion and Malthion alone are the same (.05 for each). They are both, then, "statistically significant." There is no data DDT alone. The beta coefficient was higher for malthion than the mixture, but without knowing how these are administered, it's difficult to come to any conclusions (because it's based on per kg not per house application; if malthion is more toxic or potent and is adminstered in smaller doses, a higher beta coefficient per kg means little).

Nets were the "most important" factor in reducing Eritrea's malaria problem but that's because this was used more than IRS. The authors further wrote that nets may be the most cost effective solution. They never said they were more effective or that DDT was ineffective. In fact, they said the evidence suggests the opposite, but noted environmental, economic and political obstacles to the use of DDT.

Table 8 lists significant bivariate correlations. There is nothing for DDT, so the correlation wasn't significant. Interestingly, in the multivariate analysis the coeffficient for IRS was positive (but not significant).

However, I should note that the study does refute Cosh's characterization of ITNs as essentially worthless against malaria. That's true.

But the study does not support a lot of the conclusions Tim has seemed to have drawn from it (e.g. that DDT is ineffective and even "stastically insignificant").

I would also like to respond to Ian's comment about cost above. He wrote:

"It's interesting that when presented with arguments that other pesticides are as effective as DDT, the spreaders of the blood-libel will insist that the lower price means DDT should always be used. When pesented with evidence that nets are more cost-effective than DDT spraying their reaction is "use both"."

I'm no expert on this issue but these positions are not necessarily contradictory. Any health policy is (or at least should be) based on a cost-benefit analysis. Properly understood the "costs" should include the loss of life. If there was some treatment that could prevent 80% of the population from getting malaria and it cost $1 total, then that would very cost effective. But what about the 20% of the population that are still exposed? If we include these deaths in our costs, it becomes clear what first appears to be "cost-effective" is in fact an economic disaster. Suppose there is another treatment which protects 95% of the population (all of those covered by the cheaper treament plus 15% more) but it costs $1 billion dollars. Which makes sense now? The "cost effective" treatment or the one that covers 95% of the population? Finally, there is a third option which covers 99% of the population but costs $100 trillion. Which option do you choose?

Specific to the situation: ITNs don't seem to be as effective as IRS; The study lamber linked seems to suggest this. Further, it seems that ITNs have little utility if IRS is already being used (so, one should choose one or the other, if the study is correct). The higher value of human life one uses, the more likely IRS will make economic sense.

"Table 8 lists significant bivariate correlations. There is nothing for DDT, so the correlation wasn't significant."

The table reports the results even for variables that are not significant, like you just said. So why would the exclude any DDT only data? They didn't. There isn't any DDT only data. It appears that two sprays were used, the mixture and malthion only.

Further, the authors several times note that DDT was effective and no resistance was reported (it would be odd, then, to see IRS having no statistical significance). At no place in the study to the authors suggest the ITNs are better at reducing mortality rates than IRS (in fact, they suggest the opposite). They do, however, believe that ITNs are one way to reduce the malaria epidemic and that it is arguably the most cost-effective solution.

"Interestingly, in the multivariate analysis the coeffficient for IRS was positive (but not significant)."

Interesting that you left out the fact that ITNs also showed no statistical significance in the multi-variate analysis.

Hans Gruber: You are right; Tim Lambert appears to have misread the papper and been too trusting with its authorss' results as reported. Using the data in Tables 2 and 3 of their paper, we may obtain the following multivariate regressions, first for reductions in malarial incidence as a function of the population protected by IRS and by ITN (spraying with DDT/malathion, and bednets, respectively).

CoefficientsStandard Errort StatP-valueLower 95%
Intercept22953.035081887.80753812.158567340.052242183-1033.731244
Pop. Prot. By IRS-0.0687620820.00757252-9.0804752310.069827263-0.164979662
ITN Distributed-0.0094727620.003651821-2.5939834080.234244712-0.055873346

For Pop protected by IRS/ITN dist, malaria incidence falls by .0.06876 (IRS) and .00365 (ITN)
The t-statistics are 9.08 for IRS, valid at 99%, and 2.59 for ITN, valid only at 90%, at 2 degrees of freedom.

Also using data in tables 2 and 3, multiple regression of incidence with respect to house spraying (by DDT/malathion) and ITN (bednets), we have:

CoefficientsStandard Errort Stat
Intercept20454.284713422.3767975.976631425
HOUSES SP-0.148952430.039562326-3.76500686
ITN Distributed-0.0092359820.011620443-0.794804645

i.e for each house sprayed/ITN dist, malaria incidence falls by .14895 (House Spraying) and .009 (ITN). The t-stats show 3.76 for houses sprayed, and 0.7948 for ITN bednets. Again house spraying wins!

Sadly the estimable authors of the study in question appear to have mis-reported their results in table 8, and it was indeed prima facie odd to find their "multi-variate" (sic) at odds with their simple correlations, which showed negative coefficients for both IRS and ITN. Indeed in their text they stated (p.27) that "DDT spray can eliminate up to 93-95% indoor resting density of a vulnerable vector". Needless to say that Lambert & co ignore such statements. However I do not believe in either/ or but both - 60 years ago (untreated) bednets AND spraying protected me from malaria in endemic areas of Zimbabwe (and later Mocambique), much to the regret no doubt of our genial host Tim Lambert!

Tim Curtin, I get rather different results. I don't see how you could use table 2 to malaria incidence rates since it is by zoba and not for the whole country, so I got the numbers off figure 2. Table 8 does the regression on the number of cases and not the rate, so the results I get would have to multiplied by the pop of Eritrea in 100,000s to make them match table 8. Here is the data set:

year malaria ddt malathion cumitn
2000 17 4045 2399 127863
2001 19 8362 7904 195571
2002 10 8500 5555 471609
2003 11 17423 21890 659318
2004 5 13103 30388 874070

A multiple regression:

Estimate Std. Error t value Pr(>|t|)
(Intercept) 1.991e+04 7.381e+02 26.968 0.00137 **
cumitn -3.032e-02 3.399e-03 -8.918 0.01234 *
irs 2.765e-01 6.467e-02 4.275 0.05061 .

The sign on IRS is **positive** just like they said.

Oh yeah, and bivariate on DDT:

Estimate Std. Error t value Pr(>|t|)
(Intercept) 19.0758927 5.7904554 3.294 0.0459 *
ddt -0.0006490 0.0005143 -1.262 0.2962

That's not significant.

From the op-ed page of the New York Times April 29 2006:
"Net Gains", Jeffrey D. Sachs
"The bed nets themselves cost about $6. The cost of bed net distribution per household is less than 50 cents. Including the cost of transportation from manufacturers to the villages and of follow up by trained volunteers, the total price of getting each net to the hut is under $10. Since the net lasts five years and typically two children sleep under it, the protection is about $1 per child per year. Roughly every hundred nets in use will save the life of one child a year and prevent many dozens of debilitating occurrences of malaria"

Tim Lambert:

I summed the zorbas to get national totals for the Incidence Rates in Table 2. It is reasonable to contrast the IR with the population protected by houses sprayed and with the cumulative total of ITN per Table 3. The outcomes are below, with negative and significant coefficient for IRS, less so for ITN. I think population protected by house spraying or by ITN are more appropriate inputs than physical quantities of DDT/malathion.

Pop.IRSCumITN Incidence rate
11101712786313308
2026521955718260.7
1595514716099288.8
2276756593185973.7
2594208740702820.8
Coeffs t p
Intercept 19207 20.97
Pop. Prot. By IRS -0.0498 -7.395 .0178
Cum ITN Distributed -0.00367 -2.932 .099

Kind regards

Tim

I am sorry the data in my posting got garbled, I'd be glad for a tip on how to get a Table into your Blog!

The simplest thing is to wrap the table in <pre> </pre> tags. Or use markdown and indent each table ine with four spaces.

You can't sum incidence rates to get a national rate -- you have to do a population weighted average. If you want to recreate their table 8, you need to regress on kg sprayed, though it shouldn't make much difference.

Tim Lambert:

Thanks for tip. You are right; I wrongly interpreted Table 2 as absolute numbers, when in fact it is the rates per 100,000; the rates used by you from Fig. 2 actually appear to be rates per 1,000, and not 100,000 as stated. That means you are right about the positive coeff. for DDT/malathion. There remains the inconsistency in the author's statement in their text at p.4, that both IRS and ITN were "strongly" inversely correlated with declining morbidity, and their Table 8. Another oddity is the apparent declining amount of DDT/Malathion used per house spray, from 6.2 kg in 2000 to 2.1 in 2004.

Kind regards

I am not a statistician or econometrician, but perhaps why the data makes no sense for DDT is because the substitution of malthion for ddt over time. This makes it seem that as ddt decreases the malaria rate goes down. Exclude 2004 from the results and see what you get (this is one year when ddt actually decreased). Just an idea.

Apparently the authors did not share Lambert's conclusion since they never claimed DTT was not effective (they even suggested the opposite). This is what is so odd about Lambert's assertions--they are unsupported by the study itself, as quoted above. I think they understood the limitations of the data.

The study does support Lambert insofar as Cosby claimed that nets were relatively pointless. But the study does not support Lambert's many other assertions (that ddt was statistically insignificant, etc).

Tim Curtin, maybe the declining kg per house has to do with the increase in malthion. As I noted above as a possibility, malthion may be more potent and toxic per kg and therefore they would use less mixture as the proportion of malthion increased. This would also explain the greater beta coefficient per kg for malthion than ddt.

Hans, I never said that DDt was not effective. Please take more care in your reading.

"Apparently the authors did not share Lambert's conclusion since they never claimed DTT was not effective"

If we're still talking about the Malaria Journal paper:
"Arguably the most cost effective tool in malaria prevention is the use of ITNs" is what it states.
While Cosh described it as "the study doesn't address the question of the relative cost-effectiveness of interventions."

Lambert wrote: "Hans, I never said that DDt was not effective."

Oh really? You said that DDT alone was statisically insigificant when the article didn't report any data for DDT alone. You jumped to the conclusion that Malthion was more effective because of the greater beta coefficient (per kg), but without knowledge of how it's applied that means little. You incorrectly stated that the signficance level for malthion was higher than for the mixture (actually you said correlation coefficient, but those weren't reported, so I took it to mean p value).

Sure, you never said that DDT wasn't effective generally, but you strongly suggested it wasn't effective in this case, in Eritrea. I quote:

"There wasn't a statistically significant correlation between DDT spraying and the reduction in malaria. In some places DDT is effective, in some places it isn't."

You have seemingly ignored the many times the article praised IRS spraying and DDT. You pointed to data in "table 8" that wasn't there. Then claimed it wasn't there because it wasn't statistically significant (even though the table reported variables without statistical significance). You also incorrectly claimed that Malthion was statistically significant and not the mixture.

Z wrote: "If we're still talking about the Malaria Journal paper: "Arguably the most cost effective tool in malaria prevention is the use of ITNs" is what it states. While Cosh described it as "the study doesn't address the question of the relative cost-effectiveness of interventions."

Again, whether or not it's "cost effective" is determined on what value we place on human life. If we place a low value on human life, I wouldn't be suprised if ITNs were the most cost effective solution. If we placed zero value on life, doing nothing would be the preferred solution, too. Please see my post where I (probably ineffectually) tried to describe the difficult balance between cost and efficacy.

It is I believe mistaken to treat the Eritrea data as a test of the relative efficacy of treated bednets (ITN) and residual house spraying with DDT or malathion (IRS). The project was not designed to that end, and the data as presented do not lend themselves thereto. A test would have needed to restrict ITN and IRS to different zobas rspectively with comparable malarial incidence, whereas the paper implies both were applied to some extent in each zoba. With a cumulative total of 870,000 ITN by 2004 against only a total of 259,000 people covered by IRS (about 8% of the total population), clearly ITN was the main factor in the decline in malarial morbidity. In addition the paper does not quantify the number of bednets needing to be retreated annually, while its data on DDT/malathion covers resprayed houses. The authors' negative coefficients for both IRS and ITN are valid when annual (not cumulative) totals are used.

Tim Curtin: "Again, whether or not it's "cost effective" is determined on what value we place on human life. If we place a low value on human life, I wouldn't be suprised if ITNs were the most cost effective solution."

Cost-effectiveness has nothign to do with the innate value of human life.

It has to do with comparing how much output (in this case malaria reduction)we get per dollar spent.

If you don't understand that, you understand nothing about economics.

You might also want to take a look at the value attached to a human life in soem of the more optimistic studies about the economic impact of global warming.

IIRC, oen such study argued that a human life in a devlopign country was woth around $100,000 - compred with typical values used in the economic literature of around $1,000,000.

Ian Gould: Kindly do not attribute other peoples' remarks to me; I never said what you claim I did. However cost-benefit analysis is what Hans Gruber was alluding to, not cost-effectivness. They are different animals. When you have worked that out let me know!

Ian,

Perhaps my choice of terminlogy was misleading or confusing (think of it as a cost-benefit analysis, if you want). By cost effective I was speaking about the policy's effect on society (is it costly in a comprehensive sense?) rather than the mere effect/dollar. Compare this to an environmentalist arguing that DDT isn't cost effective because of damage to the environment.

The point is that using half-measures which are arguably "cost effective" in the sense they get the greatest effect/$ may be shortsighted because it ignores the cost in human life (which can be defined in economic terms, as you point out).

I'm not sure what global warming has to do with this, but in that case costs are very high (Kyoto protocol) for uncertain and probably small decreases in global temperature. That's an interesting debate, but we should probably leave that for another time. The $1,000,000 number that is "typical" in economic lit is probably for developed rather than undeveloped countries. Productivity, like it or not, is used to come up with those numbers.

Tim, my apologies.

Ian Gould: Thanks; Hans was right, if benefits are zero then no costs should be incurred however "cost effective". The truth is that there are no known benefits of Kyoto, for high costs. Same with DDT, costs of not using are high.

Hans: "By cost effective I was speaking about the policy's effect on society (is it costly in a comprehensive sense?) rather than the mere effect/dollar. Compare this to an environmentalist arguing that DDT isn't cost effective because of damage to the environment."

Actually, Hans, environmenal economists can and do set dollar values on environmental damage all the time.

Hans: I'm not sure what global warming has to do with this, but in that case costs are very high (Kyoto protocol)

Hans, the only reason the ultra-right anti-environment "skeptics" can make their false claims about the cost-effectiveness of Kyoto is by;

a. undervaluing theh uman and environmental costs of global warming (which is where it ties in with your comments) and

b. grossly overstating the net costs.

I've repeatedly demonstrated here that the cost of Kyoto are far lower than most people think.

Ian Gould said: I've repeatedly demonstrated here that the costs of Kyoto are far lower than most people think.

I say: But that is only because those who signed up to Kyoto have done nothing to implement it.

Consider the following data on CO2 emissions from fossil fuels (source IEA Annual).

1. The share of global emissions of USA and Australia (who did not fully sign up) FELL (albeit only slightly) between the Kyoto base year of 1990 and 2003, from 24.5968% to 24.5565%.

2. The share of those who did sign fell slightly more, but only from 25.3% to 24.7%

3. The shares of China and India (who did not have to sign up) ROSE from 13% to 18%.

Thus the net effect of the signers of Kyoto on CO2 thus far is less than zero.

4. But sad to say, the Kyoto signers have actually doubled their CO2 emissions growth rate since 1990, from 0.3% pa (1980-1990) to 0.6% pa (1990-2003), while the USA and Australia have gone only from 0.37 to 0.7, and China and India continue to gallop away at around 2% pa. both before and after 1990. As Kyoto is clearly being ignored, its costs are indeed nil, as stated by Ian Gould.

Tim, go find Warwick McKibbin's ecnomic modelling of the impact on Australia of ratifying Kyoto. His conlcusion was that it would actually result in faster economic growth in the short and medium term.

Your comments abotu Kyoto bering "ignored" demonstrate your ignorance of Kyoto.

Ian Gould: perhaps there are two Warwick McKibbins! Here is what he said last year:

"It is not in the national interest to just meet the Kyoto targets and offer some subsidies. It is in the national and global interest for Australia to steer the world away from the fundamentally flawed approaches being considered."
(First published in The Australian Financial Review on February 16, 2005). If he also said what you claim, that ratifying Kyoto would boost Australia's GDP, why would he state "it is not in Australia's interest to just meet the Kyoto targets..."?

You could also read his
"The Kyoto Protocol is dead - now for the more sensible alternatives", see his Home Page.

This WK's views appear to anticipate my own!

http://www.greenhouse.gov.au/international/kyoto/pubs/modeling.pdf

Page 13: Table 7 shows that in 2010 during the first commitment period, the loss to Australia in terms of gross
national product foregone, is 0.41% of what it otherwise would have been, or roughly $A3.4 billion in
1999 dollars. This loss rises over time to 0.58% in 2015 and 0.67% in 2020. Assuming that measures
generate additional low cost reductions over the period, the loss is reduced to 0.33% in 2010, 0.47%
in 2015 and 0.51% by 2020. Finally if Australia does not participate in the Kyoto Protocol but the
rest of Annex B (excluding the United States) do participate, the cost is still high to Australia at
0.40% in 2010, 0.38% in 2015 and 0.30% by 2020. In other words the cost of participation by
Australia in Kyoto is the difference between the first row and the last row or 0.01% of GNP in 2010
rising to 0.2% in 2015 and 0.37% of GNP in 2020. This cost is reduced by introducing measures
currently identified by AGO, that reduce emissions at low cost. Indeed the scenario including
measures, is initially beneficial since it enables Australia to supply the world permit market with
additional permits at no cost, but this effect is only transitory and including measures is quickly
dominated by not participating in Kyoto by 2015.

In other words Tim, Warwick McKibbin, wh oas you note is a critic of Kyoto and who was workign for a client (the Federal government) hostile to the Kyoto accord concluded, AS I SAID, that up until 2010, the net impact of Scenario B (ratification of of Kyoto and cost-effective CO2 reduction strategies) was to increase GDP relative to the base scenario - in which Australia does not ratify Kyoto.

Of course, since that paper was written in 2002 the price of oil has roughly tripled, tell me Tim woudl you care to venture a guess as to the likely effect of higher oil prices on the NET cost of reducing oil demand?

As for Warwick McKibbin's views "anticipating" you own, Professor McKibbin is on the record as saying that urgent action needs to be taken to prevent global warming and advocating carbon taxes and emission quotas as the means to achieve that reduction.

His objections to Kyoto are specific to the particular mechanisms set up by Kyoto. In partixular, rather than allowing international trade in emission credits and allowing private companies to earn emission reduction credits which they can then sell he advocates the government selling emissions permits.

Still think his views anticipate yours?

"If we place a low value on human life, I wouldn't be suprised if ITNs were the most cost effective solution."

Given that we are discussing two options, both of whose measured outcome is reduction in malarial death rate, I don't follow your logic. One reduces the death rate by X per $ expended, the other by Y per $. Is X > Y? The actual (or estimated) value of X and Y in $ doesn't enter into it, being a common factor to both.

In any event the authors claimed, whether justified or not, that ITNs were "arguably" more cost effective; Cosh's attempt to correct his backhanded "compliment" to Africans' sophistication regarding malaria prevention programs by telling them what works and what doesn't, involves getting it wrong yet again by saying that the authors **do not** say what is more cost-effective. Hysterical blindness, seems to me.

Maybe me too, as I **still** don't see Tim L. stating that DDT is not effective; just that it is not the Holy Grail of malarial prevention in all cases in all locations at all times. Apparently for anyone to hold such an opinion constitutes a de facto ban on DDT and makes one a mass murderer worse than Hitler. I'm so ashamed.

"By cost effective I was speaking about the policy's effect on society (is it costly in a comprehensive sense?) rather than the mere effect/dollar."

?? So, you would argue that regardless of whether DDT costs more or less than the estimated $1 per child protected per year, use of treated bed nets has a more costly effect on society than use of DDT? You're going to have to do a lot more explaining to carry that off. While you're at it, you might want to post what you're using for an estimate of the cost per child protected per year, using DDT.

Ian Gould:
Here is part of Warwick McKibbin's summary of his findings(from his 2002 paper):
"When Australia is outside Kyoto, over time Australia is able to shift exports into non-Kyoto countries and therefore the GNP loss falls from 0.4% in 2010 to 0.3% by 2020. In contrast when Australia participates in Kyoto the
net costs rise over time from 0.41% of GNP in 2010 to 0.67% of GNP by 2020...."

"....The cost of participating in the Kyoto protocol to Australia by 2010 is estimated to be 0.41% of GNP or roughly $A3.4 billion in 1999 dollars. This loss rises over time to 0.58% of GNP in 2015 and 0.67% of GNP in 2020. If Australia does not participate the costs are estimated to be the same in 2010 as for the case where
Australia does participate, but only 0.38% in 2015 and 0.30% of GNP by 2020".

These quotes do not appear to match what you claimed Warwick McK said: "His conclusion was that it [ratifying Kyoto] would actually result in faster economic growth in the short and medium term."

Yes - he is saying that in scenario A in which Kyoto is ratified by the other signatories (except the US) but not by Asutralia output will fall by 0.4% compared to the baselien scenario. (It is imprtant ot note that this doesn not inply an absolute declien but a marginally slower rate of growth.)

However if Asutralia also ratifies (Scenario B), the decline compared to the baseline is only 0.3%.

sicne the other signatories have ratified regardless of Australia's position, the net effect of Australian ratification is to increase GDP in 2010 by 0.1%.

In fact, the higher oil price and the fact Australia is on track to reduce its emissions to below our Kyoto target means that the economic benefits of ratification will be higher.

The only things John Howard is doing by not ratifiying Kyoto are:

a. giving George Bush a a propaganda boost and
b. preventing Asutralian industry from selling reduction credits in the Kyoto markets.

Ian Gould: your reading ability seems as bad as your typing.

To repeat W.McK again: "When Australia is outside Kyoto, over time Australia is able to shift exports into non-Kyoto countries and therefore the GNP loss falls from 0.4% in 2010 to 0.3% by 2020. In contrast when Australia participates in Kyoto the net costs rise over time from 0.41% of GNP in 2010 to 0.67% of GNP by 2020...."

These are absolute costs in money terms: 0.67% of GNP is bigger than 0.3%. The slower rate of growth means less GNP than otherwise.

McKibbin's assumption that the Kyoto countries discriminate against Australia's exports is questionable, there appears to be no evidence for it so far, in which case there is no loss at all from not signing. What is your best estimate of the value of putative "reduction credits"? Will they be worth more than 0.41% of GNP in 2010 or roughly $A3.4 billion in 1999 dollars?

I remain perplexed by what the no-Kyoto types et al think they are preserving. Is there any question that the fossil fuel gravy train is coming to a halt, more sooner than later? And that the prudent thing to do is to start planning for this contingency? It's not as if the only thing stopping us from happily burning petroleum forever is this pesky global warming nonsense. Even in the most optimistic scenario, if global warming turns out to be a chicken-little-sky-is-falling-nothing, all that will net the industrialized countries is one final blowout before they hit the wall full speed, and if they haven't figured out how to live on an energy budget by then, it will only be worse than if they start now.

Between climate change, the encroaching end of cheap oil, the enormous balance of trade problem, which incorporates the efflux of enormous amounts of money to volatile states who generate as much human misery at home and abroad as they do energy, and the need for military intervention to mitigate both the instability of the sources of supply and their dangerous sociopolitical byproducts, exactly what glorious future do the climater change deniers think they are preserving?

The party's over folks. It was a lot of fun, but it's time to go home and get some rest, because tomorrow's going to be a tough working day.