31 August 2011

Oil Shocks and Good Times for the Global Economy

This post revisits the topic of oil prices and economic growth which was explored here earlier this year.  A new paper by Tobias Rasmussen and Agustin Roitmanfrom the IMF (here in PDF) titled "Oil Shocks in a Global Perspective: Are they Really that Bad?" which explores the largely uncharted territory of the effects of oil price increases on economic growth at the global level.  The figuer at the top shows the countries with GDPs that increase following and oil shock (those bars above the horizontal axis) and those the see a decrease (those at the left side, the US and Japan are highlighted in yellow).

Here are the paper's conclusions:
Conventional wisdom has it that oil shocks are bad for oil-importing countries. This is grounded in the experience of slumps in many advanced economies during the 1970s. It is also consistent with the large body of research on the impact of higher oil prices on the U.S. economy, although the magnitude and channels of the effect are still being debated. In this paper, we offer a global perspective on the macroeconomic impact of oil prices. In doing so, we are filling a void of research on the effects of oil prices on developing economies.

Our findings indicate that oil prices tend to be surprisingly closely associated with good times for the global economy. Indeed, we find that the United States has been somewhat of an outlier in the way that it has been negatively affected by oil price increases. Across the world, oil price shock episodes have generally not been associated with a contemporaneous decline in output but, rather, with increases in both imports and exports. There is evidence of lagged negative effects on output, particularly for OECD economies, but the magnitude has typically been small.

Controlling for global economic conditions, and thus abstracting from our finding that oil price
increases generally appear to be demand-driven, makes the impact of higher oil prices stand out more clearly. For a given level of world GDP, we do find that oil prices have a negative effect on oil-importing countries and also that cross-country differences in the magnitude of the impact depend to a large extent on the relative magnitude of oil imports. The effect is still not particularly large, however, with our estimates suggesting that a 25 percent increase in oil prices will cause a loss of real GDP in oil-importing countries of less than half of one percent, spread over 2–3 years. One likely explanation for this relatively modest impact is that part of the greater revenue accruing to oil exporters will be recycled in the form of imports or other international flows, thus contributing to keep up demand in oil-importing economies. We provide a model illustrating this effect and find supporting empirical evidence.

The finding that the negative impact of higher oil prices has generally been quite small does not mean that the effect can be ignored. Some countries have clearly been negatively affected by high oil prices. Moreover, our results do not rule out more adverse effects from a future shock that is driven largely by lower oil supply than the more demand-driven increases in oil prices that have been the norm in the last two decades. In terms of policy lessons, our findings suggest that efforts to reduce dependence on oil could help reduce the exposure to oil price shocks and hence costs associated with macroeconomic volatility.13 At the same time, given a certain level of oil imports, developing economic linkages to oil exporters could also work as a natural shock absorber.
If oil shocks are not so bad in aggregate, and associated with "good times for the global economy" then maybe the price of oil should be higher?

H/T VoxEu

Comment of the Day: The Wrong Side of History, Science and Policy

This delightful and revealing comment, apparently offered as a defense of Governor Pete Shumlin's remarks that I discussed yesterday, provides a nice capsule summary of my experiences in the climate debate.
Verbose and prolific (and cleverly snarky), most of the views expressed by this blog author are on the wrong side of history, climate science, and climate policy. I understood from multiple posts not so long ago that Dr. Pielke was going to transition to discourse on the subject of technology policy on this blog rather than climate policy. Like a moth to a flame, the content authored here remains mostly climate-based, a testament to the seduction of the defense of past positions. While the influence of environmental factors does not alone explain the causality of any individual event, most scientists agree that smoking causes cancer. You just can't pin it down to the individual cigarette. Similarly, this blog's on-going attempt to disprove linkages between GHG accumulation in the atmosphere and weather events misses the forest for the trees. The industries that require free dumping grounds in the earth's atmosphere for their profit margins must be grateful for Dr. Pielke's support, much as the tobacco companies tobacco companies loved their captured academic champions in the 1980's.
Nowadays, climate is not as much a scientific or policy issue, as it is a cultural phenomenon (read your Mike Hulme). For years I have advised my students that there is little point in doing a policy analysis of the abortion issue, as the topic was entirely political.  Perhaps one day I'll be saying the same about climate.

30 August 2011

Not Anti-Science, Just Utterly Uninformed

Yesterday, Governor Pete Shumlin of Vermont made these remarks:
I find it extraordinary that so many political leaders won’t actually talk about the relationship between climate change, fossil fuels, our continuing irrational exuberance about burning fossil fuels, in light of these storm patterns that we’ve been experiencing. Listen, since I’ve been sworn in as governor just seven months ago, I have dealt with—this is the second major disaster as a result of storms. We had storms this spring that flooded our downtowns and put us through many of the same exercises that we’re going through right now. We didn’t used to get weather patterns like this in Vermont. We didn’t get tropical storms. We didn’t get flash flooding. It wasn’t—you know, our storm patterns weren’t like Costa Rica; they were like Vermont.
A quick look at  the following paper from 2002 -- "Climate Variability and Socioeconomic Consequences of Vermont's Natural Hazards: A Historical Perspective" (here in PDF) by Lesley-Ann Dupigny-Giroux (Vermont's state climatologist), reveals this information:
One of the most pervasive hazards that impinges upon and marks the Vermont landscape is flooding. Flooding can be categorized as one of two types: flash flooding, which has a rapid onset of six hours or less from the time of the initiating event; and flooding that has a more gradual onset. Rarely does a year elapse without a flooding event of a significant magnitude being reported in at least one of Vermont’s fourteen counties or perhaps statewide, making this the number-one hazard across the state. Between 1955 and 1999, floods accounted for $16.97 million in damage annually.
And also:
[T]ropical remnants have produced widespread, and at times, catastrophic flooding. For example, the Great Flood of 1927 resulted from record rainfall totals produced by tropical storm remnants on November 3, following October precipitation totals that were already 50 percent above normal. As this decaying storm tracked directly along the spine of the Green Mountains, streams rose so rapidly that there was little time for warning. The Winooski River rose 40–45 feet above its normal level, causing land and settlement along the river to bear the brunt of the estimated $30 million in economic losses. The 1927 flood was greater than the 100-year flood on many rivers and remains today as the flood of record at many gauging stations. Eighty-four of the eighty-five fatalities during this New England-wide flood occurred in Vermont. In addition, thousands of dairy cows and other farm animals drowned. Rich topsoil on farmland either washed away or got buried under infertile silt, such that no crops could be produced for many years. Montpelier remained isolated for days and Waterbury for weeks. The flood disrupted communications across the state and with the outside world, producing a “black triangle.”
 And here is Table 1 from that paper:
Table 1: Tropical Remnants that Made Landfall In/Proximate to Vermont

Name Year Month, Day
[unnamed] 1927 November 3
Great New England 1938 September 21
#2 1949 August 29–30
Hurricane Baker 1952 September 1–2
Hurricane Carol 1954 August 31
Tropical Storm Brenda 1960 July 30
Hurricane Donna 1960 September 12
Tropical Storm Doria 1971 August 28
Hurricane Belle 1976 August 9–10
Hurricane David 1979 September 6–7
Hurricane Frederic 1979 September 14
Hurricane Gloria 1985 September 27
Tropical Storm Chris 1988 August 29
Hurricane Hugo 1989 September 22–23
Hurricane Bob 1991 August 19
Hurricane Opal 1995 October 5–6
Hurricane Bertha 1996 July 13
Hurricane Fran 1996 September 8–9
Is Governor Shumlin "anti-science" (whatever that might mean)?  No, just poorly informed.

[Thanks AS]

Certain Ignorance versus Uncertain Uncertainty

Writing in the quarterly newsletter of Risk Frontiers at Macquarie University in Sydney, Rob van den Honert has an excellent discussion (here in PDF) of the interim report of the Queensland Flood Inquiry. Some background on the topic can be found in this post from last January.

In his summary van den Honert writes of the decision of the dam operators to ignore weather forecasts of pending rainfall:
[P]redicted reservoir levels depend on expected water inflows into the dam, and that expectation would be based almost entirely upon the rainfall forecast for the catchment area. The Bureau of Meteorology supplies regular 24-hour forecasts of rainfall, and the operators also had access to the Bureau’s weather radar, even though the Bureau cautions that in some circumstances the radar can produce poor estimates, either over- or underestimating actual rainfall. Furthermore, there are far fewer rain gauges in the catchment immediately above the Wivenhoe Dam than in other areas, which means that rainfall in that area was not well recorded.

Thus [the dam operator] Seqwater claim that there were gaps in the information available on which operational decisions had to be made. This is despite Seqwater having the best rain/runoff gauge of all - the dam itself!

A 2001 Seqwater report (Feasibility of Making Pre-releases from SEQWC Reservoirs) concluded that the precipitation forecasts were not sufficiently reliable to form the basis of operational decision making for the dam. Thus this less than perfect available information was given zero weight, and not used at all to help predict reservoir levels. Effectively a “forecast” of zero rainfall was used to inform decisions about water release strategies. In other words, under the circumstances, it seems that the operators chose a scenario guaranteed to be wrong over a forecast that was likely to be uncertain.
The flood disaster arguably was exacerbated by poor decision making under flawed decision processes -- decision makers chose the certainty of ignorance over the uncertain nature of uncertainty judgments.  Indeed, as van den Honert describes the rainfall forecasts were inaccurate, but this did not mean that they would have been without value.

Ultimately, the only way that Queensland gets out of this situation will be to build sufficient water retention capacity to simultaneously meet the conflicting objectives of flood mitigation and water storage as a drought buffer. In other words, there is a technological fix here that can dramatically reduce uncertainties -- but such a strategy will cost money.

29 August 2011

A Nice Analysis of Print Media Coverage of Hurricane Irene

At the NYT FiveThirtyEight blog Nate Silver uses our normalized loss database (above) and historical loss of life data to assess the relative intensity of (mostly) print media coverage of Hurricane Irene. He finds that the coverage was in line with that of other storms since 1980. I would love to see a similar analysis exclusively focused on television news coverage -- I'd hypothesize different results.

The Folly of Emissions Trading: New Zealand and Europe

Coming Clean - New Zealand's Emissions Trading Scheme Explained from Lindsay Horner on Vimeo.

The video above was brought to my attention by a student in my graduate seminar this term.  It was done by one of his classmates in grad school in New Zealand (Thanks Adam!).  The video is exceedingly well done.  If you have 15 minutes and are interested in the debacle that is New Zealand's carbon policy, have a look.

News from Europe is similarly discouraging about the prospects for emissions trading, EurActiv reports:
European chemical manufacturers are covertly venting huge quantities of the powerful 'super greenhouse gas' HFC-23, according to a study by the Swiss Federal Laboratories for Materials Science and Technology (EMPA).

The report, published in the journal Geophysical Research Letters, says that Western Europe's emissions of HFC-23s – an 'F' or fluorinated gas mainly used as a refrigerant – are between 60-140% higher than officially reported.

Italy alone was found to be emitting 10-20 times more HFC-23s than it officially reports. The greenhouse gas has a global warming potential which is 14,800 times higher than CO2.
The UK and the Netherlands also emitted around twice as much as they claimed, although the figures for France and Germany were "within the reported values".

"We think it is scandalous," Clare Perry, a campaigner for the Environmental Investigations Agency, told EurActiv. "These gases have a very high global warming potential over a short timeline."
EMPA, the Swiss agency that conducted the research, explains the significance:
International agreements such as the Kyoto Protocol to reduce greenhouse gases (GHG) basically have one snag: it is almost impossible to independently verify whether participating countries abide by the agreement. Thus the evaluation of whether or not the countries have achieved their reduction targets is based on the official reports by the countries that are signatories to the UNFCCC (‘United Nations Framework Convention on Climate Change’). If they report reduced emissions they're sitting pretty; if not, they are pilloried.
Some might say that the response should be more regulation, more reporting, more rules -- all negotiated through a comprehensive global framework. Of course, that approach hasn't gotten very far to date.

Guest Post: Kevin Vranes on the Virginia Earthquake

NOTE: This is a guest post by Kevin Vranes.

The overwrought reaction to the M5.8 Virgina earthquake on Tuesday had a lot of native Californians snickering (ok, me included, I grew up just south of San Francisco), but I had to concede that a M5.8 in the east is not the same thing in the well-prepared west.  The fact is, the eastern seaboard has plenty of shaking in its geologic history, but little in its human history.  The frequency of very large quakes on the eastern seaboard is much longer than 1-2 generations, a period which in California seems to keep the risk fresh in the mind of locals and in the policy considerations of local governments.

Building codes in the U.S. have traditionally been highly localized, and only recently are beginning to reflect the realization that almost every state in the U.S. has some seismic risk and therefore codes should incorporate seismic design principles. Combine slow code change with old building stock (apparently the building stock is older in D.C. than in any of the 50 states - link), and you have an area at much higher relative risk to moderate shaking than California.  

The question we should be asking is, what is the true risk?  Since this is an intraplate region with lots of old, hidden faults that do not move often enough to reveal shaking risk, this is a difficult question to answer (perhaps nearing impossible in the foreseeable future).  But what about from an economic damages perspective?  In the absence of seismic data, can some measure of risk be calculated?

In 2009 Roger and I published a paper in Natural Hazards Review to examine this question. The analysis was an attempt to put historical quakes into current-day context by asking essentially this question: “If the 1906 San Francisco happened again today, with today’s population, increase in wealth and increase in damage mitigation, what would we figure the economic losses would be?”  We did this calculation for every earthquake in the U.S. since 1900 for which we could find a credible estimate of economic losses -- 80 in all.  How does this apply to east coast earthquakes, and therefore how does it reflect on Tuesday’s quake?  By giving us a measure of the frequency of quakes that cause economic damages in a certain region.

So what does our analysis tell us about Tuesday’s Virginia earthquake?  Unfortunately, the answer is: just about nothing. Since 1900 only two quakes have occurred in the east that were given damage estimates: a 1944 quake in Massena, NY (on the Canadian border, as far from NYC as you can get in New York state) and a 1954 event in Wilkes-Barre, Pennsylvania.  The Massena quake was given a $1.5M - $2M price tag, and the Wilkes-Barre event was given a $1M estimate.  Although damaging events have occurred in the mid-Atlantic and southeast in human history, none have occurred since 1900, the period for which we have good economic comparison data.

And this is the problem.  We know the major quakes can occur in the intraplate east, but they happen so infrequently and the last big ones happened so long ago, they tell us very little about what would occur today with the same shaking conditions.  The two major historic quakes in the east occurred in 1811 (New Madrid, Missouri) and 1886 (Charleston, SC).  Damages from the Charleston event was given a price tag of about $5 million in 1886 dollars.  By comparison, the 1906 San Francisco earthquake was estimated at over $500 million in 1906 dollars.

Were these quakes to occur again today damage would undoubtedly be extreme, but how extreme?  What’s left is to model with HAZUS, which can give a peek into what kind of damages you might expect from certain shaking types, and hope that your estimate for shaking risk in certain intraplate regions is fairly accurate.  Whether it is?  We’ll find out when the next big eastern seaboard quake happens near a major city.

26 August 2011

Hurricane Irene Damage Analogues?

[UPDATE August 27: This AP news story calls the ICAT Damage Estimator a "model" that "predicts" $4.7 billion in damage. Wrong (The $4.7B is the average of the  27 analogues as you can see in the image above).  The ICAT site is simply a tool to look at historical analogues and offers no predictions of the future.  The WSJ does a much better job discussing the issue.]

I've had a bunch of calls today, presumably following up from Nate Silver's post at the NYT, on potential damage from Irene.

I have used the ICAT Damage Estimator to look at all storms that fall within the spread of the various model projections (displayed above, as of 9AM MT) and here are the top  6 storms that come up.
New England Sep 21,1938 8 46,160,000,000 306,000,000 NY 2 100
Carol Aug 31,1954 16 19,240,000,000 460,000,000 NY 2 100
Agnes Jun 22,1972 18 18,880,000,000 2,000,000,000 NY TS 65
Storm 7 in 1944 Sep 14,1944 31 10,670,000,000 90,000,000 NY 1 85
Storm 7 in 1944 Sep 14,1944 36 8,320,000,000 10,000,000 NC 2 105
Storm 8 in 1933 Aug 23,1933 52 4,880,000,000 27,000,000 NC 1 80

None of the storms is really a good analogue. We should expect to see damage along the entire eastern seaboard, as well as a considerable amount of damage from inland flooding (not included in these numbers).

It wouldn't be anything more than a guess to speculate at this point on Irene's total impact, but it does seem safe to say that it's effects will be widespread and the damage total considerable.

25 August 2011

Globalization Schmobalization

A new analysis from the Federal Reserve Bank of San Francisco comes to the counter-intuitive conclusion that,
Although globalization is widely recognized these days, the U.S. economy actually remains relatively closed.
Here is part of what they find:
Obviously, if a pair of sneakers made in China costs $70 in the United States, not all of that retail price goes to the Chinese manufacturer. In fact, the bulk of the retail price pays for transportation of the sneakers in the United States, rent for the store where they are sold, profits for shareholders of the U.S. retailer, and the cost of marketing the sneakers. These costs include the salaries, wages, and benefits paid to the U.S. workers and managers who staff these operations.

Table 1 shows that, of the 11.5% of U.S. consumer spending that goes for goods and services produced abroad, 7.3% reflects the cost of imports. The remaining 4.2% goes for U.S. transportation, wholesale, and retail activities. Thus, 36% of the price U.S. consumers pay for imported goods actually goes to U.S. companies and workers.

This U.S. fraction is much higher for imports from China. Whereas goods labeled “Made in China” make up 2.7% of U.S. consumer spending, only 1.2% actually reflects the cost of the imported goods. Thus, on average, of every dollar spent on an item labeled “Made in China,” 55 cents go for services produced in the United States. In other words, the U.S. content of “Made in China” is about 55%. The fact that the U.S. content of Chinese goods is much higher than for imports as a whole is mainly due to higher retail and wholesale margins on consumer electronics and clothing than on most other goods and services.
This analysis supports the thesis of Pankaj Ghemawat that the world really isn't so "flat."  He summarizes this thesis in his challenging recent book World 3.0 which I read over the summer.  Ghemawat wrote in 2007:
I still remember a TV interview a year ago in Mumbai where the first question I was asked—quite seriously or, should I say, flatly?—was why I still thought the world was round. Spouting such attitudes—the flattening of the world, the death of distance and the disappearance of differences across countries—seems to be considered a hallmark of global thinking. But I prefer to think of it as “globaloney.”

Why? Because most types of economic activity that could be carried out within or across national borders are actually still concentrated domestically. Not convinced? Ask yourself, of all the capital being invested around the world, how much is foreign direct investment by companies outside of their home countries? Maybe you’ve heard the globaloney about “investment knowing no boundaries,” and so on. The fact is, the ratio is generally less than 10% and, while it may be pushed higher by merger waves, has never reached 20%.

As the chart below demonstrates, the actual levels of globalization associated with telephone calls, long-term migration, university enrollment, stock investment, and trade as a fraction of gross domestic product (GDP)—look at the blue bars—resemble the data presented above: they fall much closer to 10% than the levels close to 100% that one would expect if one took the gurus of globaloney at their word.

The implications here are the we should be supporting globalization -- the interconnections of markets and societies -- not pulling away.  The fact that reducing reliance on Chinese imports costs American workers more than Chinese workers is one of those inconvenient facts that will find it difficult to make it into American political discourse.

As The Economist notes of Ghemawat's analysis versus that of Thomas Friedman or Benjamin Barber:
This sober view of globalisation deserves a wide audience. But whether it will get it is another matter. This is partly because “World 3.0” is a much less exciting title than “The World is Flat” or “Jihad vs. McWorld”. And it is partly because people seem to have a natural tendency to overestimate the distance-destroying quality of technology.
The world isn't much globalized and globalization is a good thing.  Try making those arguments these days.

24 August 2011

A Democracy Working

The Economist has a great letter this week from Marc Ginzberg of Rye, NY:
SIR – The process in Congress that led to the deal to raise America’s debt ceiling was not “ludicrously irresponsible” (“Time for a double dip?”, August 6th). Extreme views were passionately expressed, policy positions maintained and pressure from party leaders withstood. But there were no shootings, no riots, no bribery, not even an appeal to the Supreme Court.

Right-wingers opposed to the deal either believed in the correctness of their position or they reflected the ideas of their constituents and want to be re-elected; what is wrong with that? If I had been a dictator, I would have imposed a very different conclusion on the debt-ceiling fracas; so, it seems, would you. But although you may regret the terms that the compromise reached, do not regret the process.
Well said!

Wednesday Climate Linkage

23 August 2011

Hurricane Irene and the ICAT Damage Estimator

The image above comes from the ICAT Damage Estimator, an online tool that lets you explore our historical database of normalized US hurricane losses.  The IDE has a neat feature which allows you to select a current storm and overlay the projected path according to the current NHC official forecast (or underlying model runs) -- The IDE can then calculate damage from all storms that made landfall within that "cone."  You can also select all storms that passed within X miles of Irene's present position and compare damage statistics on that basis.

Here are some numbers from the IDE and links to the sources:

For storms that made landfall within Irene's 5-day official forecast cone:

Average damage: ~$14 billion
Median damage: ~$3 billion

For storms that passed within 50 miles of Irene's latest position:

Average damage: ~$9.3 billion
Median damage: ~$3.6 billion

I'll provide updates this week, meantime, please explore the tool yourself - ICAT Damage Estimator.

22 August 2011

College Tuition Evolves in California

The LA Times has an article today on the changing nature of in-state tuition in California's state universities.  Here are a few interesting excerpts:
State funding for UC and the California State University was cut $650 million this year for each system. As a result, in-state UC undergrads will pay 18.3%, or nearly $1,900, more in tuition and fees than last year. Many will get financial aid, and students from families with annual incomes up to $120,000 may be eligible for reprieves on the latest increases.

UC leaders note that its tuition is on a par with other top public research schools. The University of Michigan at Ann Arbor will charge state residents about $14,000 this year; the University of Virginia, $11,600; and the University of Texas at Austin, $10,000. But UC student leaders say that living costs are much higher in California.

Sherry Lansing, a former Hollywood studio executive who chairs UC's Board of Regents, said the university was trying to find new revenue to replace state funding and avoid repeated tuition hikes. She said it must become more entrepreneurial to keep academic quality and student access.

Under consideration are efforts to boost private and alumni donations; reap more financial rewards from inventions and companies born in UC labs; and develop online classes that would attract a broader paying audience. The university also is seeking to save money by consolidating payroll and computer support functions.

"We are just at the beginning of this," Lansing said.

UC is one of many public universities experiencing a fundamental shift in how it operates, with more emphasis on revenue-producing activities, Hartle said.

"It doesn't mean you have a second-rate university," he said. "You can have a first-rate university but more focused on their bottom line. And when that happens, the public university starts looking more like the private university."
And also:
One way UC has tried to raise revenue has been to enroll more out-of-state undergraduates, who pay significantly more than California residents. The goal is for nonresidents to make up about 10% of UC undergraduates, still far fewer than at top public schools elsewhere. This fall, UC Berkeley will have the system's largest share of non-Californians, nearly 30% of freshmen. UC San Diego and UCLA are next, with about 18% each.

The University of Michigan's Ann Arbor campus, which has seen sharp state funding cuts over the last two decades, is sometimes seen as a possible model for UC.

About 40% of students in last year's freshman class at Michigan were from out of state, and the campus is boosting its recruitment worldwide. It has closed some research centers and increased alumni fundraising. Undergraduate fees vary by academic divisions, and the university's business school has moved to mainly private funding.

With such tactics, Michigan has been able to maintain financial aid for needy students and protect unprofitable academic departments, including the humanities, that have suffered deep cuts at other schools, said John Burkhardt, a Michigan education professor who directs the National Forum on Higher Education for the Public Good.

He said the campus is now a hybrid, with its public mission unchanged but its financing increasingly private.

UC is considering similar actions. The regents are debating whether campuses should charge undergraduates varying tuition levels, and UCLA's Anderson School of Management wants to wean itself from state subsidies in exchange for higher tuition.

Ink Blots, Ambiguity and Outcomes in the Real World

A fundamental problem with climate science in the public realm, as conventionally practiced by the IPCC, is the essential ink blot nature of its presentation. By "ink blot" I mean that there is literally nothing that could occur in the real world that would allow those who are skeptical of scientific claims to revise their views due to unfolding experience. That is to say, anything that occurs with respect to the climate on planet earth is "consistent with" projections made by the climate science community. Some scientists go further and argue that climate science cannot be shown to be incorrect based on experience because its projections are probabilistic. The result is that  people tend to see in climate science other things than those that can be resolved empirically -- which fosters politicization and tribal behavior.

The ink blot nature of climate science would be a non-issue if it were a field like philosophy or cosmology in which people were debating non-empirical claims for academic interests. But climate science -- or at least a very visible part of that field -- has set forth on an evangelistic path in trying to convince the unconvinced of their views among politicians and the general public.

But the ink blot nature of climate science leaves climate scientists in a position of arguing from authority or demanding that people need "trust us."  The typical mode of engagement with skeptics by many visible climate scientists is to argue how right they are (and wrong/evil the skeptics are) -- but what skeptics need instead is to hear what it would mean for climate scientists to be wrong. If one cannot be wrong, then experience cannot be used to adjudicate claims. (I am aware of various debates that have occurred about using variables such tropical tropospheric temperature trends, ocean heat content, water vapor feedback, etc. in an effort to falsify claims of climate science.  While I am by no means an expert on these debates my understanding is that the climate science community argues that uncertainties/variability are so large as to make such claims not inconsistent with their views, taking us back to square one.)

Consider by way of example how the field of economics handles such situations.  Ratings agencies issue ratings associated with the likelihood of default for lenders, and utilize a language very similar to that found in the IPCC. For instance, here is how S&P defines their credit ratings (PDF):
In our view, likelihood of default is the centerpiece of creditworthiness. That means likelihood of default--encompassing both capacity and willingness to pay--is the single most important factor in our assessment of the creditworthiness of an issuer or an obligation. Therefore, consistent with our goal of achieving a rank ordering of creditworthiness, higher ratings on issuers and obligations reflect our expectation that the rated issuer or obligation should default less frequently than issuers and obligations with lower ratings, all other things being equal.

Although we emphasize the rank ordering of default likelihood, we do not view the rating categories solely in relative terms. We associate each successively higher rating category with the ability to withstand successively more stressful economic environments, which we view as less likely to occur. We associate issuers and obligations rated in the highest categories with the ability to withstand extreme or severe stress in absolute terms without defaulting. Conversely, we associate issuers and obligations rated in lower categories with vulnerability to mild or modest stress.
It is generally understand that the ratings agencies were wrong in their estimates of the likelihood of default among mortgage based securities. How do we know this?  Well, the global economy melted down for one.

I am aware of no one who has claimed that the ratings of subprime mortgages cannot be judged wrong simply because ratings are based on likelihood estimates. But this is exactly where some climate scientists would find themselves, if they were arguing about economics rather than climate.

It is not just the subprime crisis either where experience matters for the evaluation of expectations. Recently, Paul Krugman discusses the recent S&P downgrade of the US as follows:
When assessing the downgrade, the question of track record comes up. As I understand it, countries that defaulted in the past were almost always downgraded well before the default happened; but in all such cases, the markets were already signalling big trouble before the rating agencies moved.

The question should be, in cases when the markets aren’t signalling worry but the agencies downgrade anyway, how often are they right?

The answer, I believe, is never — not for Japan 2002, not for various European countries in the late 1990s, not for Canada 1994.
Until the climate science community steps out from behind academic parsing and hiding behind uncertainties, it will continue to be an ink blot, and one that many people evaluate using political and other non-scientific criteria.

There are two ways for the climate science community to move beyond an ink blot (if it wishes to do so). One would be to advance predictions that are in fact conventionally falsifiable (or otherwise able to be evaluated) based on experience.  This would mean risking being wrong, like economists do all the time. The second would be to openly admit that uncertainties are so large that such predictions are not in the offing. This would neither diminish the case for action on climate change nor the standing of climate science, in fact it may just have the opposite effect. 

The default will be the status quo, which means climate science as inkblot -- and the associated arguments from authority, "trust us" and politicization that comes along with it.

21 August 2011

More Fun With Uncertainty Guidance

Here is a another math exercise.  In its AR4 report the IPCC says:
The uncertainty guidance provided for the Fourth Assessment Report draws, for the first time, a careful distinction between levels of confidence in scientific understanding and the likelihoods of specific results. This allows authors to express high confidence that an event is extremely unlikely (e.g., rolling a dice twice and getting a six both times), as well as high confidence that an event is about as likely as not (e.g., a tossed coin coming up heads). Confidence and likelihood as used here are distinct concepts but are often linked in practice.
Here are some specific definitions to help you answer some questions.

A. "high confidence" means "about 8 out of 10 chance of being correct".
B. "extremely unlikely" means "less than 5% probability" of the event or outcome
C. "as likely as not" means "33 to 66% probability" of the event or outcome

So here are your questions:

1. If the IPCC says of a die that it has -- "high confidence that an event is extremely unlikely (e.g., rolling a dice twice and getting a six both times)" -- how should a decision maker interpret this statement in terms of the probability of two sixes being rolled on the next two rolls of the die?

2. If the IPCC says of a die that it has -- "high confidence that an event is about as likely as not (e.g., a tossed coin coming up heads)" -- how should a decision maker interpret this statement in terms of the probability of a head appearing on the next coin flip?

Please provide quantitative answers to 1 and 2, show your work.

20 August 2011

Fun With Epistemic and Aleatory Uncertainties

Deep in the comments on an earlier thread Paul Baer offers the following hypothetical:
In my statistics class, I ask my students "what is the probability that when I flip this coin, it will land heads." (And yes, assume it's a fair coin.)

Of course they answer 50% or some equivalent.

Then I flip it and hold it covered on the back of my hand. Then I ask, "What is the probability that this coin is heads." There's usually some puzzlement. Someone says "50%". And I say, "but either it's heads or it isn't. How can there be a fifty percent chance it's heads?"

Then I ask "what odds would you give me if I bet that it's not heads?" Eventually those who know what betting odds mean understand the point. Even when something has happened (like, the deck has been shuffled and the card that will be dealt could be known under some epistemic conditions DIFFERENT FROM OURS) we have to ACT as if the odds are, well, what we think they are.
To which I responded:
Consider the following case:

You flip a coin in your class and ask for the probability of a head. A savvy student replies:

[S1]: The odds of a head are 50-50

You then reveal to the class that the coin is not fair, in fact there is a 75% chance of a tail. You ask the student, now what are the odds of a head? (All while the flipped coin sits on your hand)

The student now replies:

[S2] The odds of a head are 25%.

Q1. Now would it be fair to say that [S1] was incorrect?
Answers gladly accepted.

19 August 2011

Friday Funny - Deniers Risk Alien Attack

This from an academic paper by researchers at Penn State and NASA (here in PDF, note that ETI = extraterrestrial intelligence and METI = messages to ETI):
Humanity may just now be entering the period in which its rapid civilizational expansion could be detected by an ETI because our expansion is changing the composition of Earth’s atmosphere (e.g. via greenhouse gas emissions), which therefore changes the spectral signature of Earth. While it is difficult to estimate the likelihood of this scenario, it should at a minimum give us pause as we evaluate our expansive tendencies.

It is worth noting that there is some precedent for harmful universalism within humanity. This precedent is most apparent within universalist ethics that place intrinsic value on ecosystems. Human civilization affects ecosystems so strongly that some ecologists now often refer to this epoch of Earth’s history as the anthropocene [79]. If one’s goal is to maximize ecosystem flourishing, then perhaps it would be better if humanity did not exist, or at least if it existed in significantly reduced form. Indeed, there are some humans who have advanced precisely this argument [80-82]. If it is possible for at least some humans to advocate harm to their own civilization by drawing upon universalist ethical principles, then it is at a minimum plausible that ETI could advocate harm to humanity following similar principles.

The possibility of harmful contact with ETI suggests that we may use some caution for METI. Given that we have already altered our environment in ways that may viewed as unethical by universalist ETI, it may be prudent to avoid sending any message that shows evidence of our negative environmental impact. The chemical composition of Earth’s atmosphere over recent time may be a poor choice for a message because it would show a rapid accumulation of carbon dioxide from human activity. Likewise, any message that indicates of widespread loss of biodiversity or rapid rates of expansion may be dangerous if received by such universalist ETI. On the other hand, advanced ETI may already know about our rapid environmental impact by listening to leaked electromagnetic signals or observing changes in Earth’s spectral signature. In this case, it might be prudent for any message we send to avoid denying our environmental impact so as to avoid the ETI catching us in a lie.

18 August 2011

CPR Colorado Matters Interview on In-State Tuition Reform

Colorado Matters on Colorado Public Radio has a lengthy and meaty interview with me on the subject of in-state tuition reform.  The interview concludes with me explaining how difficult it is for universities to consider change or entertain new ideas.  In related news, CPR reports:
A spokesman for the Colorado Board of Regents says the regents have not discussed this idea and have no plans to do so.
Have a listen and please return with any comments, criticisms or questions!

17 August 2011

Obituary: John Marburger

Nature asked me to write an obituary for John H. "Jack" Marburger, III who died in late July.  It was a challenging and humbling assignment.  It appears in this week's issue (also here in PDF).

Marburger should remembered not only through what others write about him, but for what he himself wrote.  To that end, below I republish a wonderful speech of his that I came across in my research for the Nature piece. The speech was delivered in July, 1981 upon Marburger being announced as president of New York's Stony Brook University.  Enjoy.

John H. Marburger 3d made these remarks in a speech at his recent inauguration as third president of the State University at Stony Brook in July, 1981.

There are two kinds of ignorance: the kind removable by education, and the other kind, which is defined by the limits of current knowledge. It seems to me that in extolling the virtues of higher education we have overemphasized the removable ignorance and encouraged the notion that more is knowable than is actually the case. This has mischievous consequences.

Living with ignorance is for an academic something like living in sin. We are supposed to conquer ignorance through research, and urge our students not to be satisfied until they understand what is happening around them. Modern education consists of continual exposure to the knowable domain of human experience. We profess humility and declare the limitations of our knowledge, but spend all the time in our lectures talking about what is known. That may be inevitable. The consequence, however, is that our students and we ourselves, I am afraid, form the habit of assuming that things can be explained.

Our conviction of knowability surely derives from the success of the physical sciences, where nature allows herself to be mimicked accurately by mathematical models. The predictive success of science has been so great that efforts have been made in every other practical field to introduce scientific methods. The results have been useful, but reliable predictions can be achieved outside physical science only in the simplest situations.

Even where we do not have a clear understanding of the relation between means and ends, however, we use language patterned after the more successful sciences to describe events. This encourages the illusion, among the inexpert, that we know more than we do. Medicine has employed this practice with success for millennia.

Even our admiration for directness and clarity of thought reinforces the habit of the assumption of knowability. Events are simplified by electronic journalism to ''problems'' expressed in language that suggests both their origin and their solution. The practices of encapsulation, of briefing and of interviewing encourage oversimplification and enhance the illusion of comprehensibility.

What is wrong with glossing over this other ignorance? At the very least, it increases the impatience and frustration that we always feel when things do not go smoothly. If we are supervisors or managers or taxpayers, we tend to expect more from our organizations or our employees or our governments than is reasonable.

When something goes wrong, our first impulse is to blame it on poor planning or on the ignorance of those responsible. If the ignorance is of the removable kind, blame and censure are justified. But if it is the other kind of ignorance, the inevitable kind, then censure is inappropriate. Failure to appreciate the distinction between avoidable and unavoidable ignorance leads to unrealistic management practices. It is a hallmark of inexperienced managers.

In times of peace and prosperity, defects in our world view are of little consequence. Adversity shows them up. When funds are scarce, budget controllers attempt to match allocations as closely as possible to estimated needs. If they believe that those needs can be estimated exactly, they will have little patience with managers who protest ignorance of how to do it. Each legislative act, each regulation, is a hypothesis about the relation between a desired end and a means to achieve it. To the extent that hypothesis is incorrect, the end cannot justify the means.

Budget control, legislation and regulation are processes that depend very sensitively on our understanding of cause and effect in human affairs. For reasons that I have suggested, these processes tend to assume greater understanding than we possess. They tend to mistrust protestations of ignorance and to punish inability to control events even when they are uncontrollable. In our democratic society, some of the blame must rest with voters and taxpayers, who compare our successes in scientific ventures with our failures in economic and social affairs.

The inevitablility of ignorance is not necessarily cause for despair. We do have ways of managing our affairs that accommodate uncertainty. As physical science has been the model for mechanistic views of human affairs, engineering has provided models that admit ignorance and chaos. Engineering thoughtfulness about the problem of communicating in the presence of random disturbances and the problem of unattended operation of devices in unpredictable environments suggests ways of approaching administration, law and social reform. Modern trends in management theory do exploit these notions.

In the final analysis, decisions about how to act are made by individual men and women. They may be central planners or local managers. They may be aware or unaware of their ''immense'' ignorance. But they all possess an instrument that has been found by experience to be extremely powerful in dealing with ambiguity and surprise: the human mind.

We do not know how the mind works, how it transforms information into action. We do not understand intuition, or wisdom, or sound judgment. But such qualities do exist, and they are universally admired. However we choose to do our business in state or school or home, we must not embrace a course that limits the application of these qualities.

We do not understand human societies, or what motivates them to war or work together. We do not understand precisely the relation between acts of individuals and their consequences in the larger community. We do know that communal cooperation is normal and that individuals are often enormously influential in organizing and focusing the efforts of society.

Our best chance for coping with the reality of ignorance is to rely upon the vast integrative power of the human mind. We must learn to develop this power in ourselves and to recognize it in others, even if we do not understand it. And we need to respect it and to organize our affairs so that it may be brought to bear in all situations illdefined and poorly understood. The kind of mental development I have in mind is not simply instruction in various systems, theories or models of how things and people work, but also exposure to the quandaries of the real world and how real people have responded to them in the past.

In this respect, the part of university education whose content (not whose presentation) is the least methodical seems to be of the greatest value. The view that I am advocating is a humanistic one, because it recognizes explicitly human capacities that cannot be duplicated or replaced by systems, policies or machines. And it is precisely the humanistic material in our curricula that seems best suited for developing those capacities.

We have come to this point by a long argument, but I can find none better for the value of the humanities in modern education. The humanities are valuable because they deal openly with the inevitability of ignorance and the consequences thereof. They show us how great men and women faced incomprehensible situations. They tune the instrument by which ultimately we all grapple with the question of how to act without sufficient knowledge. And they urge us to free that instrument, the educated human mind, from the restraints of ignorance, even ignorance of ignorance itself.

Academic Exercises and Real World Commitments

My exchange last week with James Annan, a climate modeler, was interesting for several reasons, not least, for his suggestion that the projections of the IPCC could not be judged to be wrong because of their probabilistic nature. This view strains common sense, to put it mildly.

Here I argue that our disagreement lies not in different views about the nuts and bolts of probabilistic forecasting, but rather our views on whether the IPCC is engaged in providing guidance to decision makers about the probable course of the future, or instead, is engaged in an academic exercise.  This would seem to be a natural point of disagreement between an academic involved in modeling and a policy scholar. James does the climate science community no favors by personalizing the debate, so here I'll stick to the issues, which are worth a discussion.

Let’s start with an analogy to make this distinction more clear.  Please note that I reject James choice of a die roll as an analogy because it begins with a presumption that the probabilities for various future outcomes are in fact known with certainty. While such an assumption is highly complementary to climate modelers, the fact is that the actual probability distributions for the future are unknown.  That is why evaluation of probabilistic statements is necessary.

Let’s say that I make the following statement:
[A] It is very likely that a team from the AFC will win the 2012 Super Bowl.
You, being a Green Bay Packer fan, take issue with my statement and ask me if I want to bet, but first you ask me what I mean by “very likely.” I explain that by “very likely” I mean at least a 90% chance. You then ask if I will give 9 to 1 odds on a $1 bet on the Super Bowl. Since I am confident in my statement (and I believe that the bet gives me good odds as I think that the chances could even be higher than 90%), I agree to bet and we shake hands.

Consider the following way that events might play out -- Let’s flash forward to the Super Bowl in 2012. Imagine that the Denver Broncos (hey, it is my example;-) beat the Packers. In that case, I would win the bet and you would pay me $1. We would both agree that the expectation expressed in my statement [A] was correct in a common sense understanding of the term. Had the Packers won the game, I would part with $9 and we’d agree that my expectation was incorrect.

Now let’s consider an alternative ending to this scenario. Let’s assume that the Packers win the game. I immediately protest to you that I actually did not lose the bet -- at least not yet, because my statement was probabilistic in the sense that it referred to the entire set of potential Super Bowl games in which I judged that the AFC would win 90% or more of them. It just so happens that that this was one of those cases that happens to fall in the 10% of possible outcomes. In order to judge the winner of the bet we would have to replay the Super Bowl many times, and I suggest that we do so on Madden 2012, a computer-game simulation.

You would reply, “Huh? Whatever, you lost the bet, where’s my $9?”

The difference between the two endings lies in whether one views a probabilistic statement as a basis for committing resources for a future that will play out in only one way or as an academic exercise to be explored using simulations of the real world.

On the one hand, in the first case, a Packers’ victory would mean that the expressed judgment in [A] turned out to form the basis for a losing bet. It is important to understand that the judgment [A] may have been perfectly sound and defensible at the time that it was made – perhaps it was based on my expertise or perhaps I actually ran a Madden 2012 simulation 10,000 times under various scenarios to create an ensemble projection space as the basis for my judgment. Perhaps then the outcome was just bad luck, meaning that the 10% is realized 10% of the time. Actually, we can never know the answer to whether the expectation was actually sound or not, only that a commitment based on the judgment lead to a loss of $9.

On the other hand, if I actually meant statement [A] simply as an academic exercise, perhaps in relation to some simulation of the real world, I should not be betting in the real world.

Let’s continue with the analogy and say I was a sports handicapper and you wanted to get a sense of my skill in giving odds on outcomes. One way that you might evaluate my performance is to look at all of my Super Bowl predictions over time. But that doesn’t offer much of an experiential basis with which to judge probabilistic projections. Another way that you might judge my performance is to look across my handicapping of many sports and see how the collective outcome compares with what I have projected. For instance, if I make 100 projections that are judged to be “very likely” you will have a lot of confidence in my skill if 90 of those 100 projections actually realize, even if they are across 100 different, incommensurable events.  But even then there would be problems in assessing my skill (e.g., what if there are 10,000 sports handicappers and you come to me because my record is better than the others, is that because I am good or just lucky?  But I digress).

Now let’s relate this back to climate science. As an example, the IPCC says the following:
It is very likely that hot extremes, heat waves and heavy precipitation events will continue to become more frequent.
If decision makers commit resources (let’s assume intelligently, fully recognizing and understanding a 90% probability) based on this projection, and it turns out that the actual future experienced on planet Earth is one of constant or declining hot extremes, heat waves and heavy precipitation events, then the common sense view will no doubt be that the IPCC got this one wrong.

Climate modelers who protest such a common sense view based on what they describe as the impossibility of verification of probabilistic statements generated from ensemble projections from modeled simulations of the real world will be laughed off as out-of-touch academics, and rightly so. Infallibility doesn't even work for the Pope.

16 August 2011

In-State Tuition Reform Continued

[UPDATE: The AM760 interview is now online here.]

Today I'll be on AM 760 Colorado Progressive Talk Radio today at 9AM to discuss the issue of in-state tuition reform.  Later, I'll be on Colorado Matters on Colorado Public Radio (I will update with broadcast time when I know it).  Who knew that in-state tuition reform was such a hot topic?

In the many reactions I received yesterday to the front page Camera article on my recent CHE piece-- ranging from emails to comments at the bus stop -- it is clear that most people fail to distinguish between (a) the sticker price of tuition and (b) the state subsidy of tuition. What this means is that people really have no idea that tuition for in-state residents is actually subsidized, but at a level that does not allow the books to balance. It also means that some Colorado residents, when they hear a call for a levelized tuition for in-state and out-of-state students, think that the cost to them will double.  This is not necessarily the case.
Have a look at the graph above (source), which shows Colorado state funding per resident student.  In round numbers, the state provides about $3,000 per student.  If the state allows the university to charge about $8,000 per student in tuition, and the cost of running the university requires a break-even tuition of about $14,000, then you can see that the result is a loss of $3,000 per in-state student.  That adds up fast -- it is a  $60 million shortfall.

How does the university make up the difference?  Answer: Out-of-state students!  In 2010 the proportion of out-of-state students at CU was larger than at any time since 1975 (data, it was 34.5% in fall, 2010 and state law allows it to go to as much as 45%).  From 2005 to 2010 the number of CU in-state undergraduates shrunk by 1% while the number from out-of-state increased by 17%.

If the state were to allow a levelized tuition, then it could still subsidize the costs of attending CU for state residents -- at whatever rate it deemed appropriate by subsidizing directly those residents accepted to the university. Imagine what would happen if the university charged the full cost of tuition ($14,000) and the state provided individual students with a check for its current subsidy per student ($2,800) -- Colorado residents would immediately see that the subsidy does not go very far, and it might lead to a different conversation about the state role in higher education. In addition, subsidizing the student rather than the institution would also allow for a greater focus on need-based support rather than simple geographic residency. Of course, there would be no requirement for each state university to follow the same tuition model, a diversity of approaches would probably make sense,
It is worth saying something about cutting costs. Some might argue that the university should simply cut $60 million of fat from its budget in order to balance the books. There are two responses to this. One is that all the fat has been cut and the university is now actually deep into bone (see, e.g., graph above). The second is that while academia is far from a perfect market, at some level you do get what you pay for -- consider that Stanford charges about $38,000 per year in tuition. CU simply cannot compete with the Stanfords of the nation by charging one third as much in tuition. Now perhaps Colorado residents or its legislature do not want CU competing with the nation's top research schools for students, faculty, research, etc. -- an open debate about where the state schools (all 12 of them) ought to aspire to in a national context would be a valuable contribution to the discussion.

The bottom line is that the current model for state university support is badly broken in a number of ways.  University performance is disconnected from its market price. Admissions decisions are increasingly based on who can pay the most.  Budgets have been cut to the bone. Ultimately, something has to give, and in the end it will be the quality of education and research at the institution. For those wishing for a cheap university, be careful what you wish for, as you might just get it.

14 August 2011

Who Says Universities Aren't Conservative Places?

Universities do not embrace change. Sometimes they don't even embrace talking about change.  The Boulder Daily Camera has done a short news article on my recent proposal in the Chronicle of Higher Education that the State of Colorado eliminate in-state tuition for its flagship university(ies).  Reached by a reporter my university offered this official response:
Boulder campus spokesman Bronson Hilliard said that while the idea is interesting, CU officials aren't considering it. There are too many constraints -- such as a state law that requires CU's freshman class to be made up of 55 percent in-state students, averaged over a three-year period.

"If it were as easy to do as he posits in his piece, we probably would have done it years ago," Hilliard said.
Stymied by the law. Of course, it is precisely that legal requirement that I'd like to see changed, so invoking it as an obstacle to change is really to miss the point. And there is no shortage of discussions about changing Colorado law by university officials -- every year CU officials are complaining about Colorado laws that have led to a reduction in State support. I guess those laws are OK to discuss changing.

At the same time officials at CU and CSU are not shy about the funding incentives that motivate them to look to out-of-state students as cash cows:
CU, where the mix of students is 55 percent Colorado residents and 45 percent nonresidents, is also working to increase the number of international students, who pay nonresident tuition and fees that will total about $29,000 to $31,000 for the 2011-2012 school year, compared to about $8,000 to $12,000 for state residents.

[CU President Bruce] Benson says the added revenue could total about $80 million for the university, after expenses such as English as a Second Language classes. He predicts the students will come from England, where recent tuition increases led to riots, as well as Saudi Arabia, China and South America. Other schools are looking worldwide for higher-paying students. Colorado State University, where 80 percent of students are from Colorado, is building relationships with institutions in China.

"It certainly helps the bottom line when we have a few nonresident students," says Rick Miranda, CSU's provost and executive vice president. "Certainly the trend is to be more global no matter what sector you are playing in."

CSU undergraduate residents pay about $7,000 in tuition and fees, while nonresidents pay about $23,000. Miranda says CSU does some targeted recruiting in California, Texas, Chicago and Minnesota. "We are spurred on to be energetic about our recruiting when our revenue streams are in jeopardy," he says.
 Change is difficult at universities, who ever said that they are not conservative? ;-)  As the Camera article alludes, change is already coming to the in-state tuition model, and it won't be the most conservative campuses that secure the benefits of that change.

11 August 2011

How Many Findings of the IPCC AR4 WG I are Incorrect? Answer: 28%

I suspect that headline will raise some eyebrows.

In a paper just out in Climatic Change today Rachael Jonassen and I perform a quantitative analysis of all 2,744 findings found in the three 2007 assessment reports of the Intergovernmental Panel on Climate Change.  Here is the abstract of our paper:
Jonassen, R. and R. Pielke, Jr., 2011. Improving conveyance of uncertainties in the findings of the IPCC, Climatic Change, 9 August, 0165-0009:1-9, http://dx.doi.org/10.1007/s10584-011-0185-7.

Abstract Authors of the Intergovernmental Panel on Climate Change (IPCC) Fourth Assessment Report (AR4) received guidance on reporting understanding, certainty and/or confidence in findings using a common language, to better communicate with decision makers. However, a review of the IPCC conducted by the InterAcademy Council (2010) found that “the guidance was not consistently followed in AR4, leading to unnecessary errors . . . the guidance was often applied to statements that are so vague they cannot be falsified. In these cases the impression was often left, quite incorrectly, that a substantive finding was being presented.” Our comprehensive and quantitative analysis of findings and associated uncertainty in the AR4 supports the IAC findings and suggests opportunities for improvement in future assessments.
The paper characterizes the various findings of the report in terms of the uncertainty guidance used by the IPCC.  The paper includes various summary statistics and discussion.

The answer to the provocative title of this post is found in the following part of the paper:
If we confine our attention to those findings that refer to the future, one can ask how many IPCC findings can be expected to become verified ultimately as being accurate? For example, if we consider findings that refer to future events with likelihood in the ‘likely’ class (i.e., >66% likelihood) then if these judgments are well calibrated then it would be appropriate to conclude that as many as a third can be expected to not occur. More generally, of the 360 findings reported in the full text of WG1 across all likelihood categories and presented with associated measures of likelihood (i.e., those summarized in Table 2 below), then based on the judgments of likelihood associated with each statement we should logically expect that about 100 of these findings (~28%) will at some point be overturned. 
A footnote to this paragraph explains: "This calculation assumes that each finding can be treated independently. If the findings are not independent (e.g., they are cumulative) then this calculation would result in a higher estimate."  Since this is just mathematics following from the IPCC uncertainty guidance, it should be obvious, but appears never to have been actually calculated.

What does it mean?  Nothing too interesting, really -- science evolves and any assessment is a snapshot of knowledge in time. However, I suspect that some people will get excited or defensive to learn that by the IPCC's own logic, the report's future-looking findings could include 28% or more that will not stand the test of time. Of course, such excitement and defense are part of the context that the IPCC and its critics have together created, which has led to incentives to hold the IPCC up as some sort of sacred text or to denigrate it as a sham. Our work suggests neither. Instead, from the perspective of its assessment products it is a valuable if imperfect organization.

Our paper concludes:
Although the IPCC has made enormous contributions and set an important example for global assessment of a vexing problem of immense ramifications, there remain clear opportunities for improvement in documenting findings and specifying uncertainties. We recommend more care in the definition and determination of uncertainty, more clarity in identifying and presenting findings and a more systematic approach in the entire process, especially from assessment to assessment. We also suggest an independent, dedicated group to monitor the process, evaluate findings as they are presented and track their fate. This would include tracking the relationship of findings and attendant uncertainties that pass up the hierarchy of documents within AR5. Strict rules for expressing uncertainty in findings that are derived from (possibly multiple) other findings are needed (see, e.g., the second example in the Supplementary Material).

It is not the purpose of this note to discuss other, related scientific assessments of climate change knowledge; but, we do note that our preliminary analysis of the U.S. Global Change Research Program Synthesis and Assessment Products suggests a far less systematic application of the guidance supplied to authors of those documents and far less consistent application of the defined terms. We believe that the concerns we have expressed here, and the resulting recommendations, apply more broadly than the IPCC process.
You can find the full text here.  Our full dataset of IPCC findings is online here.  Comments welcomed!