As others have done before me, I am honored to receive the APSA John Gaus Award. As I prepared this lecture, I realized that the Gaus award has been given by APSA 26 times; mine is the 27th. The first was awarded to Herbert Kaufman whose work set a very high standard for this honor. Reviewing the list of the other Gaus award recipients provides a picture of the development of our field. It includes a variety of individuals who represent different approaches to the intersection of public administration and political science. Among the recipients are seven individuals who had a major and personal influence on my work: Aaron Wildavsky, Frank Rourke, George Frederickson, Martha Derthick, Lou Gawthrop, Larry Lynn, and David Rosenbloom. Others are people who have been important to my own intellectual development.

Climate change represents one of our greatest public policy challenges. A broad, well-established and international scientific consensus exists that our planet is undergoing climate change. The question is not whether there is climate change, but how best to respond to it. Climate change is a global phenomenon that requires a global effort unlike anything previously attempted. This global challenge is complicated by related and more controversial questions about causal responsibility. Although convincing evidence shows that climate change is a result of human behavior, much less agreement exists on how this should factor into policy.

We are at a turning point when it comes to the political implications of climate change. Given the reality of a future in a climate-changed world, it is time for us—broadly as a species, but particularly as academics—to move beyond the foci of the last few decades on the politics of preventing climate change through global agreements. There is a growing literature on the obvious need to slow the impacts of climate change, develop postcarbon energy systems, and design new forms of global environmental governance. Beyond these immediate needs, however, climate change poses a range of new problems and requires a broader research agenda for a climate-challenged politics.

Scholars of justice theory often refuse to apply their principles to concrete social or political issues; instead, they develop those principles in abstraction from contemporary value conflicts or policy debates, preferring to remain silent on how justice might inform controversial political decisions. Rawls, for example, casts questions concerning the application of his “justice as fairness” conception across national, generational, and species boundaries as among the several “problems of extension” for which his theory may or may not be equipped, noting that “the idea of political justice does not cover everything, nor should we expect it to” (1993, 20–21). Even where he applies his justice theory to problems of international relations in his The Law of Peoples, Rawls describes its application as merely “an extension of a liberal conception of justice for a domestic regime to a Society of Peoples” (1999, 9), as though constructing and applying justice principles are entirely discrete steps, with its application to concrete social or political issues a unidirectional project of wielding static principles as practical tools, offering nothing of importance to a normative theory's development.

Geoengineering, the “deliberate, large-scale manipulation of the planetary environment in order to counteract anthropogenic climate change” (Shepherd et al. 2009, 1), is attracting increasing interest. As well as the Royal Society, various scientific and government organizations have produced reports on the potential and challenge of geoengineering as a potential strategy, alongside mitigation and adaptation, to avoid the vast human and environmental costs that climate change is thought to bring (Blackstock et al. 2009; GAO 2010; Long et al. 2011; Rickels et al. 2011). “Geoengineering” covers a diverse range of proposals conventionally divided into carbon dioxide removal (CDR) proposals and solar radiation management (SRM) proposals. This article argues that “geoengineering” should not be regarded as a third category of response to climate change, but should be disaggregated. Technically, CDR and SRM are quite different and discussing them together under the rubric of geoengineering can give the impression that all the technologies in the two categories of response always raise similar challenges and political issues when this is not necessarily the case. However, CDR and SRM should not be completely subsumed into the preexisting categories of mitigation and adaptation. Instead, they can be regarded as two parts of a five-part continuum of responses to climate change. To make this case, the first section of this article discusses whether geoengineering is distinctive, and the second situates CDR and SRM in relation to other responses to climate change.

Geoengineering has been roughly defined as “the intentional manipulation of planetary systems at a global scale” (Keith 2000; Schelling 1996). This definition is neither as precise nor as informative as some would like. Nevertheless, we can fix ideas by focusing on the most prominent current proposal, which is to inject sulfate aerosols into the stratosphere to deflect incoming radiation and so cool the Earth's surface. This is a paradigm case: if anything counts as geoengineering, stratospheric sulfate injection (hereafter SSI) does.

Climate change is confirmed by a global scientific consensus. Although no serious disagreement about whether climate change exists, deep controversy remains about what should be done about its many harmful effects. These effects are wide-ranging and include, for example, the increasing threat to coastal wetlands from rising sea levels, the greater likelihood of droughts harming agricultural production, and the spread of tropical diseases (Pachauri and Reisinger 2008). One further effect is the increasing risk of triggering an environmental catastrophe, such as an ice age, that might result in major loss of human lives.

In January of 2012, the Primary Model predicted that Barack Obama would defeat Mitt Romney by a comfortable margin. He did. Obama's popular vote total topped Romney's by more than four million votes. The Primary Model's forecast was by far the earliest of those assembled in the special 2012 PS symposium (45 (4): 614–617). Even so it predicted an Obama victory with greater certainty (almost 90%) than any other one. To put this performance in perspective, note that several models got it wrong in 2012, and hardly any of the other ones picking Obama to win did so with much certainty. Many preelection polls also failed to predict an Obama victory with their final estimate, including the venerable Gallup Poll. How did the Primary Model get it right so far in advance of Election Day 2012?

The October 2012 issue of PS published a symposium of presidential and congressional forecasts made in the months leading up to the election. In the following articles, the forecasters assess the accuracy of their models.

The Time for Change Model once again correctly predicted the winner of the US presidential election. In late August, before the Republican and Democratic national conventions, the model predicted that President Obama would win 50.6% of the major party vote to 49.4% for his Republican challenger, Mitt Romney. When all the votes are counted, it now appears that Obama will end up with approximately 51.8% of the vote, making the Time for Change Model one of the nation's most accurate statistical forecasting models. Despite the fact that this prediction was made more than two months before Election Day, the Time for Change forecast was more accurate than the results of many national polls conducted immediately before the election including the Gallup and Rasmussen tracking polls both of which predicted a popular vote win for Mitt Romney.

On August 1, 2012, we prepared a forecast of the 2012 presidential vote for PS. Our model contains two variables: (1) the cumulated weighted growth in leading economic indicators (LEI) through quarter 13 of the current presidential term and (2) the incumbent party candidate's share in the most recent trial-heat polls, which were for the month of July. What mostly distinguishes our model from others is the reliance on leading indicators from the quarter ending in March of the election year. The early reading of LEI works well as a predictor because it summarizes growth in the economy leading up to the election year and also provides advance indication of changes in the economy during the election year. The exact equation and the exact forecast change as the poll readings change during the election year.

Our Proxy Model of presidential election forecasting declared, from data issued six months before the November contest, that Obama would garner 52.7% of the two-party popular vote. [See the model release in our Monkey Cage blog-post (Lewis-Beck and Tien 2012a) on September 18, 2012, and in the October issue of PS (Lewis-Beck and Tien 2012b).] Thus, that forecast called the correct winner, with a point estimate error of only 0.9 percentage points.

On September 10, 2012, immediately following the close of the Democratic Party's national nominating convention and 57 days before Election Day on November 6, my Convention Bump and Economy Model predicted that Barack Obama was likely to receive 51.3% of the national two-party popular vote. The Convention Bump and Economy Model consists of Gallup's preconvention preference poll, the net convention bump in the polls, and an adjusted second quarter GDP growth rate. The forecast pegged the certainty of an Obama plurality at 67%. No sure thing, but more likely than not. The traditional Trial-Heat and Economy Model predicted Obama's vote at 52.0%. The Convention Bump and Economy Model's prediction was the preferred forecast in 2012 because of the lateness of the parties' nominating conventions. Democrats did not even begin their convention until after Labor Day.

President Obama received approximately 51.5% of the two-party vote in the 2012 election. The last Bread and Peace Model forecast of Obama's vote share, based on advance estimates of 2012:quarter 3 personal income posted on October 26, 2012, by the Bureau of Economic Analysis (BEA), was 46.6%—lower than the 47.5% forecast appearing in the October issue of PS, which was based on July 27, 2012, BEA data. The Bread and Peace Model therefore underpredicted Obama's vote by 4 to 5 percentage points, equivalent to around 2 model standard errors. The president's vote therefore benefited from a +2-sigma composite shock to Bread and Peace Model fundamentals. Figure 1 shows actual and predicted values for 2012 in perspective of incumbent vote shares at all presidential elections 1952–2012.

The Incumbency and National Conditions (presidential approval and aggregated personal finances) Model predicted President Obama would garner 47.9% of the two-party vote, whereas he ended up with 51.8% (based on available information on December 3, 2012). The error in this forecast (3.9 points) is somewhat higher than the average out-of-sample error from 1952 to 2008 (2.4 points). Although the forecast was off the mark, the addition of the 2012 result to the data set does little to change the slope estimates, and the overall fit of the model is only slightly worse (table 1).

At its most basic level, this forecast of the 2012 presidential election performed successfully. The model forecasted the reelection of President Obama. With 53.8% of the two-party vote; he ended up with 51.8% (as of November 28, 2012), yielding an error of 2.0 percentage points. Prior to 2012 the average out-of-sample forecasting error was 3.34 points. Here, the model does better than the average of the previous years by almost a point. Running the model with the added data point does not change matters appreciably. The economic item increases slightly and the logged time in the White House item weakens slightly. The overall fit of the model is just a bit lower.

The Fiscal Model of presidential elections not only failed to predict that President Obama would win reelection handily, but the error incurred in the forecast of the incumbent share of the two-party vote was such (46.9% vs. 51.8% or more) as to warrant a rating of “inaccurate” or “quite inaccurate” in Campbell's table of benchmarks for evaluating forecasts (Campbell 2012, 611). This recap is a preliminary assessment of what went wrong.

Perhaps the major story in forecasting the 2012 election is the growing awareness of the benefits of aggregating multiple sources of data for improving prediction. Most prominently, polling analysts including Simon Jackman, Drew Linzer, and Nate Silver made the strong case, ultimately validated by the election results, that combining information from multiple polls gives a better picture of the electorate than any poll analyzed in isolation.

For the second presidential election in a row the prediction of the “Klarner Model” was within half a percent of the actual result. This model forecast that Obama would receive 51.3% of the two-party vote, while he actually received 51.8%. The model also called all states correctly, with the exception of Florida, which was predicted to be narrowly lost by Obama with 49.7% of the vote. These forecasts were made on July 15, 2012. This success does not indicate the model is better than the other models in the symposium as luck is a major determinant of which model gets closest to the mark. Although the Klarner Model correctly called 49 out of 50 states, that means little in an election where merely calling states on the basis of 2008 results would have called all but two correctly.

One hundred and forty two days before the 2012 US presidential election our final State-by-State Political-Economy Model gave an advantage to Barack Obama with 51.6% of the popular vote (error margin ± 4.47) and 324 electoral votes (Jerôme and Jerôme-Speziari 2012). On November 6, 2012, with 51.6% of the vote and 332 electoral votes, the Democratic incumbent wins a second term. Regarding certainty of an Obama plurality, the model gave a probability of victory by 64%. In 2012, it seems that this was enough to ensure a good predictability.