Imagine contemporary political economy as the electromagnetic spectrum. At one end are very long-run processes, such as the emergence of the political institutions that generate divergence between rich and poor countries in the world economy. At the other end are short-run phenomena, such as the movements in financial markets that we ponder while reading the stock pages with our morning coffee. For the last 10 years, William Bernhard and David Leblang have been exploring this latter end of the spectrum with clarity and rigor, and between them have pretty much defined the standards for the field.

Democratic Processes and Financial Markets gathers together, but also integrates and extends, a series of essays that the two have published in major journals on the politics of financial markets. The underlying theory is simple: Investors incorporate expectations about political developments into their portfolio decisions. When market actors face uncertainty, they hedge against risk by moving into instruments that are less vulnerable to adverse political outcomes.

Actually testing this proposition is complicated, however, because well-functioning markets quickly incorporate and discount predictable political outcomes: The election of Margaret Thatcher in 1979 should have different effects than the election of George W. Bush in 2000 or Angela Merkel in 2005. Moreover—and here is where the empirical strategy becomes particularly complex—we can only test any of these political propositions against an appropriately specified baseline of what the markets would have done in any case.

In taking on this challenge, Bernhard and Leblang produce some of the most truly interdisciplinary political economy done by political scientists to date. The book considers the effects of politics on exchange rates, stock and bond prices, and interest rates. Depending on the outcome they seek to explain, they draw on tools from contemporary financial economics—predictions generated by futures market prices, capital asset pricing models, covered interest arbitrage models—to establish an appropriate baseline from which to measure the perturbations caused by politics.

But what do we mean by “politics”? The authors begin the book with an exercise in demolition. They run through a series of explanations that have been tested in the literature, including the presence of elections per se, incumbent partisanship, partisan change, the decisiveness of elections, and institutional factors such as the nature of exchange rate commitments. They find that this standard list has surprisingly weak explanatory power with respect to exchange rate movements. The reasons go back to the informational approach underlying the book: These factors do not necessarily provide new information to markets, and as a result we should not expect them to have effect. We need to capture, rather, those periods characterized by genuine political uncertainty.

To undertake this task requires both a theoretical intuition and a method for testing it. The theoretical bet placed by a book subtitled Pricing Politics owes a large debt to Laver and Shepsle's work on cabinet formation in parliamentary systems. The method relies on breaking up the political timeline of the advanced industrial democracies into periods that can be explored in more detail using high-frequency financial data, down to the day in some cases: the period prior to an election, the period during which the new government is being negotiated, the period immediately following the sitting of the new cabinet.

Following Laver and Shepsle, it is in the period when the cabinet and the allocation of portfolios is being negotiated that the potential for genuine uncertainty arises. Dominant or strong parties dampen uncertainty; in such settings, the effects of elections and changes in government should be muted. But the absence of such parties makes the outcome of the government-formation process much less predictable. It is in those moments when we would expect investors to respond by hedging risk, by fleeing stocks for bonds, for example (Chapter 3). Similarly, they find that increased risk of cabinet dissolution creates uncertainty, affecting interest rates (Chapter 4).

The authors seem almost apologetic about their more focused work on Austria, New Zealand, the United States, and Britain (Chapters 5–7). But in some ways, these chapters—pursuing a broadly similar logic—provide an even more appropriate laboratory for the careful and sophisticated time-series econometric work that characterizes the book throughout. For example, in their chapter on Britain, they consider the possibility that exchange rate movements might affect the political popularity or even the survival of government, thus generating recursive political economy processes that they explore using sophisticated econometric models from financial economics.

This book is going to generate a lot of theses, and should be adopted in doctoral seminars on political economy; it deserves its own week as an introduction to the modeling of financial markets in the short run. I hope for the authors' sake it crosses over into the financial community, as analysts think through how political information might be modeled more explicitly in making portfolio decisions. The exposition is wonderfully clear: Bernhard and Leblang walk through the theory, the empirical modeling, and the results with great clarity and humility, saying what they can say from the data and not overreaching their conclusions. This is really a kind of gold standard for work on the relationship between politics and markets, a book not just summarizing a strand of work but creating it de novo.

So what is not to like (beyond the jealousy of wishing you had done such neat and systematic work)? One concern is with the theory. The focus throughout is on one particular source of political uncertainty, almost relentlessly so. As a student of the developing world, where institutions are more fluid, I wished the authors would get away from cabinets for a moment and consider a richer menu of challenges. Politicians dominate the political landscape here; sources of uncertainty, from adverse trends in product markets, from developments in the private sector (think Enron), or from social forces such as labor or ethnic conflict, receive less attention, if any at all. These should rightly be seen as extensions of the approach, and by no means in conflict with it, but they would provide a more rounded and complex picture.

A more serious complaint has to do with the meaning of it all. Short-run economic processes are certainly worth modeling; they are the very stuff of markets. But that investors do not like uncertainty, and respond by hedging risk, is hardly counterintuitive. When Bernhard and Leblang step off their core message—government formation and dissolution and uncertainty—the conclusions they draw are less compelling. They note, for example, that their results are consistent with a model of politics in which politicians are forced to respond to market forces through either economic or political reform. In fact, this can in no way be drawn from their work, as they admit; it requires an altogether different level of analysis complementary to, but ultimately distinct from, the focus on the very short run. They close with a nice discussion of what we do not know, including a plea for still more microlevel detail. For example, we know surprisingly little at the behavioral level on how individuals actually process political information in making financial decisions. But this approach—however interesting—will not solve the aggregation problem: how you get from short-run market behavior to policy choices. Bernhard and Leblang know well—and other parts of their work show—that those problems require an exploration of other frequencies along the electromagnetic spectrum.