Democratic Processes and Financial Markets: Pricing Politics

by William Bernhard, David Leblang

Overview

The authors examine the conditions under which democratic events, including elections, cabinet formations, and government dissolutions, affect asset markets. Where these events have less predictable outcomes, market returns are depressed and volatility increases. In contrast, where market actors can forecast the result, returns do not exhibit any unusual behavior. Further, political expectations condition how markets respond to the political process. When news causes market actors to update their political beliefs, market actors reallocate their portfolios, and overall market behavior changes. To measure political information, Professors Bernhard and Leblang employ sophisticated models of the political process. They draw on a variety of models of market behavior, including the efficient markets hypothesis, capital asset pricing model, and arbitrage pricing theory, to trace the impact of political events on currency, stock, and bond markets. The analysis will appeal to academics, graduate students, and advanced undergraduates across political science, economics, and finance.

Product Details

ISBN-13:	9780521678384
Publisher:	Cambridge University Press
Publication date:	07/24/2006
Edition description:	First Edition
Pages:	272
Product dimensions:	6.02(w) x 9.02(h) x 0.59(d)

About the Author

William Bernhard is Associate Professor of Political Science at the University of Illinois at Urbana-Champaign. His work has appeared in the American Journal of Political Science, the American Political Science Review, International Organization, International Studies Quarterly, the Journal of Politics, and the Quarterly Journal of Political Science.

David Leblang is Associate Professor of Political Science and Research Associate at the Institute for Behavioral Sciences at the University of Colorado. He previously taught at the University of North Texas and at the College of William and Mary. His work has appeared in The American Journal of Political Science, the Quarterly Journal of Political Science, The International Journal of Finance and Economics, and International Organization and International Studies Quarterly. William Bernhard and David Leblang have received the Franklin L. Burdett Pi Sigma Alpha Award from the American Political Science Association and the Robert H. Durr Award from the Midwest Political Science Association.

Read an Excerpt

Cambridge University Press
978-0-521-86122-9 - Democratic Processes and Financial Markets: Pricing Politics - by William Bernhard and David Leblang
Excerpt

---

Introduction

Political processes – elections, cabinet formations, referenda, legislative debates – determine a government's economic policies, which, in turn, condition the environment for investment. By anticipating the results of these processes, savvy investors can re-allocate their portfolios to meet a changed policy context. Asset owners, therefore, have a keen interest in predicting political outcomes: Which party will win the election? Who will form the government? What will be the new government's policy priorities?

The collective decisions of investors, in turn, shape how markets respond to political processes. Whether markets react to political events in a systematic manner remains an open issue. Do asset prices behave differently when left parties are in office? Will the election of a particular candidate cause a run on the currency? Does divided government – where the executive comes from a different party than a majority of legislators – cause a market downturn?

The consequences of these market responses extend far beyond questions of portfolio allocation. The investment decisions of asset owners fundamentally shape economic performance. Shifts in asset allocation can sustain an economic upswing or cut-off growth. With the technological and financial integration of asset markets,price movements can cascade across borders and markets, creating a world-wide boom or a systemic crisis.

Changes in market conditions have distributional consequences as well. Economic actors reliant on the stock market for income – firms, pensioners, etc. – are vulnerable not only to falling equity prices but also to volatility in market returns. Individuals or firms employed in export oriented and import competing industries see their fortunes rise and fall with fluctuations in the exchange rate. And changes in the market for government bonds may place public expenditures at risk if governments have to pay a higher rate to borrow.

Asset markets can also exert tremendous influence on political processes. Sharp price shifts, for instance, can threaten the approval ratings of incumbent politicians. Declining asset prices may signal government weakness. Increases in the costs of borrowing limit fiscal policy flexibility, preventing politicians from funding new policy initiatives or forcing them to cut popular programs.

Indeed, some political economists contend that the nature of capital markets threatens the quality of democracy by limiting the choices available to voters. Owners of mobile capital may react to “unfavorable” election outcomes by shifting their assets out of a market or even out of the country, punishing voters for choosing politicians unfriendly to their interests. As global markets become increasingly integrated and market actors have an ever larger range of cross-border investment options, the pressures on politicians to provide a pro-market policy environment intensify. To ensure continued investment – and continued economic growth – politicians must favor the interests of capitalists over the policy demands of other constituents. In anticipation of a potential market reaction, therefore, politicians may adopt policies and institutions to benefit asset owners. Voters, too, may expect such a market shift and alter their vote choice away from their own preferred party to one that will cater to the needs of the market. Asset owners, therefore, compel governments to adopt policies and institutions that favor their interests, even without taking any overt action.¹

While political economists have identified the potential implications of larger, more integrated capital markets for democracy, they have not closely investigated the mechanisms that connect political events and market behavior. In fact, markets respond to political events in a variety of ways. In some instances, markets react calmly to political changes. In others, political events touch off frenetic market activity. Without a better understanding of the links between politics and asset markets, we cannot draw strong conclusions about the consequences of financial market integration for democratic governance.

In this book, we explain and empirically evaluate the conditions under which political events affect asset markets. We demonstrate that the political uncertainty created by democratic political competition can induce sharp changes in currency, bond, and equity market behavior, potentially hurting economic activity. We also draw out the political consequences of this financial market behavior. Turmoil in financial markets can have political costs, reducing public approval and increasing the costs of borrowing. Our results, therefore, can help explain recent institutional reforms in the industrial democracies. With deeper levels of financial market integration, both owners of financial assets and politicians have an incentive to support institutional reforms to (1) reduce political uncertainty and (2) limit potential financial market volatility.

INVESTMENT PORTFOLIOS AND FINANCIAL MARKETS

Financial markets, where economic actors trade financial assets such as bonds, stocks, currency, and real estate, are indispensable to a well-functioning economy.² These markets allow economic actors – individuals, firms, and/or governments – access to pools of capital by matching borrowers and lenders. Borrowers can forgo future consumption for present consumption by borrowing today and repaying tomorrow with interest. Lenders give up immediate consumption in the hopes that future consumption will be enhanced by making a profit on their investment.

In these markets, assets are bought and sold at different prices. Two factors condition the price behavior of these assets. First, the price is a function of the demand for and supply of capital. When little demand for (or large supply of) an asset exists, the price will be low. Strong demand and limited supply, on the other hand, result in higher prices.

Prices also reflect the risk associated with the holding of an asset. The pay-off from owning an asset often occurs in the future. Consequently, the expected return may not be realized. Risk is a relative, rather than absolute, concept: we can only compare the risk of two assets at a single point in time or the risk of a single asset across two points in time. Typically, investors require compensation for holding a risky asset. For instance, a stock of a start-up company may pay a slightly higher rate of return than a blue-chip stock because the risk of the company going bankrupt is higher.

The variety of financial markets provides economic actors with the ability to transfer wealth and risk across time and space. They also allow investors to hold a diversified portfolio of assets. Diversification permits investors to spread or reduce the overall risk associated with delaying consumption and investing. Understanding how individual investors balance risk and return should, in theory, yield clues about overall market performance. Indeed, the major models of finance, the capital asset pricing model, arbitrage pricing theory, etc., begin with assumptions about the micro-behavior of individual investors to generate predictions about aggregate market behavior. We discuss these models in subsequent chapters.

Investors can diversify across a variety of assets and markets. We focus on three in this book: bonds, stocks, and foreign exchange. Bonds represent promises made by the borrower to the lender that a fixed amount of capital, with interest, will be repaid at the time of maturity. Some bonds also periodically pay a fixed amount of interest between the time of issue and the time of maturity. Both corporations and governments issue debt instruments in the form of bonds to finance long-and short-term expenditures and to counteract shortfalls in revenue.

Corporations and firms can raise funds by issuing stock or equity. For investors, equity instruments provide an ownership interest in the firm and the promise of a share of the company's profits if there are any to be had. Ownership and profits are proportional to the amount of the firm's stock owned by the investor.

Economic actors may also trade in foreign exchange markets. These markets allow individuals and firms to diversify risk by purchasing and holding currency issued by foreign governments. Foreign exchange is traded largely in two different types of markets: spot and forward markets. In spot markets, currency is bought and sold today for immediate delivery. Forward markets, on the other hand, allow a contract to be signed today at a given price for delivery of the currency at some point in the future. Forward currency markets are especially important to firms engaged in cross-border transactions as they provide an opportunity to hedge their currency risk/exposure when future changes in the spot rate of a currency are unknown.

Stock, bond, and currency markets are broadly accessible to individuals, firms, and governments. And they are deep: trading in these assets reaches well over one hundred billion dollars a day. Most importantly, we can observe changing behavior in these markets on a daily – and in some cases on an even more disaggregated – basis, making them the perfect laboratory within which to evaluate the impact of democratic politics.

EXPLAINING MARKET REACTIONS TO POLITICAL EVENTS

A small interdisciplinary literature explicitly examines how democratic political processes – elections, cabinet negotiations, etc. – affect financial markets.³ These studies reveal considerable variation in how markets respond to political events. Attempts to explain this variation have, for the most part, centered on institutions and partisanship.

Some political economists argue that the configuration of political and economic institutions conditions how investors will perceive the impact of a potential political change.⁴ Institutional commitments that insulate economic policy from the direct control of elected politicians enhance policy stability, reassuring market actors that policy will not drastically change in the event of a partisan shift in the composition of the government. Therefore, institutions that limit policy discretion, like an independent central bank or an exchange rate commitment, should diminish market reactions to political events. Others argue that institutions that produce policy continuity will reduce market volatility. Some contend that proportional representation electoral systems, where legislative seats are distributed to parties in proportion to their vote totals, are more likely to limit policy fluctuations than winner-take-all majoritarian systems and, therefore, will dampen price shifts.

A second set of explanations centers on partisanship.⁵ These arguments assume that parties differ in their policy priorities, based on the interests of their constituents. Right parties appeal to middle-and upper class voters and emphasize price stability and fiscal balance. Left parties, drawing support from the working classes, place more weight on employment and redistribution. These policy objectives suggest that asset owners prefer right parties to left parties. Political economists have evaluated whether the partisanship of the incumbent, a change in the partisan composition of the government, and the direction of partisan change systematically move markets.

Attempts to link institutions and partisanship to variations in market behavior have had, at best, mixed empirical support. For every paper finding that, say, an independent central bank limits exchange rate volatility or that right parties strengthen stock market performance, there is another paper showing that the independence of the central bank is irrelevant to currency market behavior and that stock markets rally when left parties are in power.⁶ Surveying the literature on market behavior surrounding political events, few consistent patterns emerge.

The failure of simple hypotheses underscores the need for more theoretical development about the conditions under which democratic politics affect market behavior. We argue that previous explanations fail to model adequately the information available to market actors. While institutions and partisan cues provide a context for the information available to market actors, they do not capture the dynamic flow of information during a political process. Thus, these studies are unable to capture how market actors form and update their expectations in response to political developments.

POLITICAL INFORMATION AND MARKET BEHAVIOR

Given the mixed findings of previous research, it is tempting to conclude that political events have little systematic effect on market behavior. But the strong intuition that politics matters for markets remains. The widespread reporting of political events in the business-related news media implies that markets do respond to politics. Sharp shifts in market behavior during several electoral campaigns provide anecdotal support. Why, then, are the results of academic research so inconsistent?

We contend that the mixed results reflect the difficulties of mastering different disciplines. Too often, political scientists investigate asset markets without taking advantage of the extensive research in economics and finance on the behavior of these markets. (Indeed, our first paper on the topic came back from the review process with a rather unsubtle suggestion that we learn something about how markets actually work before putting anything in writing.) On the other hand, financial economists often come at the problem with a naïve and under-developed understanding of political phenomena. Only by combining the two sets of research can we evaluate how markets respond to politics. This book, therefore, takes an interdisciplinary approach, drawing theoretical insights and empirical strategies from political science, finance, and economics. Using sophisticated models from both political science and finance, we hope to introduce more rigor and precision to how political economists study capital markets.

As with many analyses of asset market behavior, we build from the assumption that market actors engage in economic activity in efficient markets.⁷ The efficient markets hypothesis (EMH), simply defined, states that asset prices immediately and fully reflect all available and relevant information (Fama 1991).⁸ As soon as information hits the market, it is incorporated into asset prices. Market actors, therefore, cannot make excess profits (relative to the risk(s) they have undertaken) by utilizing trading strategies based exclusively on publicly available information.

Economists distinguish different types of market efficiency based on the information sets available to market actors. The weak form of efficiency assumes that past prices cannot be used to make abnormal returns (profits). That is, current prices are a function of information related to past prices and trading volumes. The semistrong form of efficiency means that public information cannot be used to make abnormal profits – in other words, prices reflect all publicly available and relevant information. Finally, the strong form of efficiency presupposes that prices reflect all relevant information at all times, including private information (e.g., insider trading).⁹

The idea of market efficiency is closely related to the statistical model of a “random walk.” A variable follows a random walk if changes from the present value of the variable occur randomly, although these random changes are drawn from a known distribution. The efficient markets hypothesis implies that asset price changes follow a random walk – that is, changes in asset prices will occur randomly since asset prices already reflect the relevant information available to traders. Publicly available and predictable information, therefore, cannot systematically influence an asset's price.

This does not mean that information never affects asset prices. News – that is, the arrival of unanticipated information – may influence the trading behavior of economic agents. The publication of earnings reports, macroeconomic announcements, and interest rate changes may cause buyers and sellers to reevaluate their trading priorities if the information comes as a surprise. Although asset owners may know when the information will be announced and are likely to have a guess about substance of the announcement, as long as they do not know for certain the actual content, then the release of that information can change prices.

These arguments apply to political information as well. Information about political processes is widely available. The news media regularly report the activities of political leaders, offer public opinion surveys, and speculate on eventual policy outcomes. Party labels and endorsements provide other sources of information. The profusion of political information allows asset owners to form expectations about the likelihood of different political outcomes: who will win election, whether the bill will pass, or whether the country will go to war. Investors use this information to shift their portfolios to balance risk and return. This information about political processes will be quickly incorporated into asset prices.¹⁰

In the short-term, democratic political outcomes are often predictable. It is usually possible to forecast when the prime minister will call for elections, who will be elected, and how cabinet portfolios will be allocated, etc. before the event actually occurs. The predictability of political events allows asset owners to shift their portfolios in anticipation of the eventual outcome. Consequently, markets may not react to the actual resolution of the political event since economic actors had already anticipated the outcome. Instead, market adjustment may have occurred prior to the event when the likely outcome became apparent. The predictability of so many political processes, therefore, helps explain why there is often little market reaction to political events.

In situations where political outcomes are less predictable, however, we are more likely to observe markets responding to political events. If the eventual outcome remains unpredictable throughout the process, market actors will shift their assets only once the event has occurred. Consequently, market reactions are likely to be sharper in response to the conclusion of an unpredictable process than when market actors can anticipate the outcome.

But markets are also likely to behave differently during unpredictable political processes, not just at their conclusion, as traders shift their assets in response to new information. During these periods, investors may shift their portfolios toward assets that are insulated from the policy consequences of a political change, producing a drop in the price of assets that are more vulnerable to political influence and an increase in the price of the more insulated asset. For instance, asset holders may shift from bonds into equities (or vice versa) or substitute foreign assets for domestic ones.

One way to observe the market response to political events is to evaluate whether asset returns are “abnormal.” The idea of an abnormal return reflects that, in the absence of a shock like an earnings announcement or a political event, forecasts of asset returns are usually fairly accurate. By comparing the actual return with the forecast return, it is possible to evaluate how market actors shifted their assets in response to an event.

Political processes, however, may not necessarily generate changes in the mean behavior of asset prices. The unpredictability of political outcomes may instead contribute to increased market volatility – a measure of the predictability of asset prices. Higher volatility indicates that price forecasts are less certain.¹¹ The variation in expectations surrounding the eventual political outcome may cause higher volatility, where prices jump around more than in periods where political outcomes are more predictable.

We develop these arguments and empirically evaluate their implications for a variety of markets. The importance of political information in shaping market behavior demands careful consideration of measurement issues. Unfortunately, current work in political economy tends to operate at a high level of institutional aggregation, employing simple indicators of electoral systems, partisanship, or policy institutions. These indicators are unlikely to capture the flow of information available to market actors. Moreover, these institutional indicators are static. Markets learn about politics. We need to be able to update the information available to market actors during a political process. To examine the impact of politics on market behavior, therefore, we need a better understanding of (1) the type of political information that is available to market actors and (2) how that information is processed.

We contend that political science theories can provide insight into the political information available to market actors about the partisan and policy consequences of political events. Political scientists have developed models about equilibrium behavior in electoral competition (e.g., Downs 1957; Cox 1987, 1989; Shepsle 1991), cabinet formation (e.g., Laver and Shepsle 1996; Schofield 1992; Riker 1962), cabinet dissolutions (e.g., Lupia and Strom 1995; Smith 2004) and policy choice (e.g., Krehbiel 1998; Cameron 2000). Where a predictable equilibrium exists, we argue that markets should be able to anticipate that outcome and adjust accordingly. In the absence of a predictable equilibrium, however, economic actors may be less able to forecast the consequences of a political event, creating increased market volatility or abnormal returns. Using these more nuanced models of the political process helps determine the political information actually available to markets.

We also consider how political information is processed. Prior beliefs condition how the arrival of information affects the expectations of market actors. In some instances, information may simply reinforce the market's prior beliefs about the likely result of a political process, producing no shifts in asset allocation. In contrast, unexpected political developments may force market actors to update their beliefs about the eventual outcome, producing a shift in market behavior. We seek to model how market actors update information during the political process.

RESEARCH STRATEGY

We evaluate the impact of democratic political events on asset market behavior using a sample of the industrial democracies from 1980 through 2003. This choice not only reflects the availability of the political data necessary to measure market expectations – for example, polling information, party platforms, etc. – but also allows us to take advantage of advanced models of democratic politics, particularly for parliamentary systems. Further, the focus on “normal” politics in established democracies should bias our tests against finding evidence that politics influences markets. In these countries, extraordinary political events such as a coup d'etat or a major government default are rare. Asset owners, therefore, can have confidence that political events will not cause a major disruption to markets. By looking at countries where property rights are secure, we are able to isolate the role of political predictability in shaping market behavior.

Our sample period coincides with the globalization of financial capital. During this time, major changes in the regulation of domestic and international capital markets throughout the industrialized democracies, combined with technological innovations, led to an explosion in the pace and volume of asset trading.¹² The increasing integration of financial markets during this period should again make it difficult to find evidence that political events affect asset behavior.

Our concern with developing a more sophisticated understanding of the information available to markets also implies an empirical strategy. In order to capture the political information available to market actors, we must analyze specific events. Only by investigating the arrival of political information at identifiable times can we assess the relationship between political processes and market behavior. Therefore, we complement our aggregate analyses of markets and politics across space and time with studies of particular cabinet formations and elections. While one might argue that these “case studies” lack generalizability, we view them as necessary to illustrate the mechanisms of political and market equilibration. Moreover, the case studies are carefully grounded within a larger theoretical framework, improving our ability to use them as the basis for more general inferences.

The size and integration of financial markets dictates that we work at the lowest level of temporal aggregation possible. Often, studies of asset price behavior use monthly, quarterly, or even annual data to gauge how politics affects markets. But financial markets are so integrated and responsive that any evidence of a political influence on the behavior of asset prices is likely to dissipate rapidly. This necessitates that we work with temporally disaggregated data. For the most part we employ daily or weekly political and financial data series. In some cases, we work with “tick” data, which documents each transaction in a market.

We draw on a variety of models of market behavior from the finance literature, including the efficient markets hypothesis (EMH), capital asset pricing model (CAPM), and arbitrage pricing theory (APT), to trace the impact of political events on the behavior of currency, stock, and bond markets. We do not evaluate whether these models provide the best explanation of market behavior. Rather, these models provide a clear analytic benchmark against which to measure market responses to political events. In each chapter, we employ one of these models to analyze behavior of a particular market, offering a short introduction to the main assumptions and implications of the model, a characterization of the market under investigation, and a discussion of how we expect to observe the effects of political information. We also utilize a variety of empirical tests and methodological tools developed by financial economists to evaluate the arguments. Each chapter, therefore, can be read individually.

This “multiple markets, multiple tests” approach will disappoint readers looking for strict adherence to contemporary guidelines of research design in political science. But the approach allows us to gain leverage on the main research question in a variety of settings. Similar results across different markets or with different methods provide strong evidence in favor of the argument. Moreover, the approach permits us to discuss different markets, theories, and techniques. By doing so, we hope to encourage other political economists to address new ideas, adopt new tools and, most importantly, ask new questions.

OUTLINE

The next four chapters investigate the effect of political uncertainty on financial markets, arguing that political predictability leads to more stable market behavior. Chapter 2 demonstrates how democratic political events – elections and cabinet dissolutions – affect the efficiency of currency markets by focusing on the relationship between the spot and forward exchange rate. The efficient markets hypothesis implies that the forward exchange rate – the price of the currency deliverable 30 days in the future – should be an unbiased predictor of the future spot exchange rate. Using weekly data from eight countries, we demonstrate that the forward exchange rate is a biased predictor of exchange rate changes more often during periods of potential political change than when the government's tenure in office is secure. That is, these political events appear to make currency markets less efficient. Nevertheless, simple hypotheses based on incumbent partisanship, partisan change, electoral institutions, and exchange rate commitments cannot account for the variation of market reactions to different political events.

The inability of these simple hypotheses to account for variations in market behavior points to the need to develop an alternative approach to evaluate the conditions under which democratic politics affect market behavior. Chapter 3, therefore, draws on political science theory to examine the proposition that unpredictable political processes affect markets differently than events with predictable outcomes. We test the argument by examining the impact of cabinet formations in parliamentary democracies on stock and bond markets. Political scientists have developed models to predict the results of a cabinet formation process. We draw on work by Laver and Shepsle (1996) who assume that parties bargain over the distribution of cabinet portfolios, rather than over the cabinet's policy priorities. From this assumption, they develop a spatial model that predicts the outcomes of cabinet negotiations based on the distribution of legislative seats, the policy positions of the parties, and the salience of different issue dimensions in that system. Using their model, we determine whether a “strong” party exists – i.e., one party that is in an advantageous negotiation position – in over 70 cabinet formations between 1970 and 2003. When a strong party exists, coalition negotiations should be fairly predictable. In these cases, we find no evidence of abnormal stock market returns during the cabinet formation process – that is, the returns are consistent with the market's performance in the months immediately prior to the election. When a strong party does not exist, however, the outcome of coalition bargaining is less predictable. In these cases, stock returns are depressed during the negotiation period, suggesting that investors have re-allocated their portfolios and increased their holdings of assets with less risky returns.

Financial markets are increasingly integrated. Shifts in asset market behavior, therefore, are likely to spill across borders. Chapter 4 investigates how political events in foreign countries affect domestic equity and bond markets. We first estimate the size of these politically induced foreign shocks. We then examine the institutional determinants of their size, including monetary arrangements, exchange rate regimes, and electoral systems. Independent central banks, fixed exchange rates, and predictable politics inhibit the transfer of politically induced financial shocks across countries. These institutional arrangements are similar to many adopted in the industrial democracies in the past twenty years, particularly in the European Union. We return to the issue of institutional reform in the conclusion.

Chapter 5 examines how political events affect interest rates. Higher interest rates can inhibit the flexibility of fiscal policy, preventing politicians from initiating new programs or forcing them to cut popular ones. Higher interest rates may also dampen economic activity in the private sector. Incumbents may be held electorally accountable for slower growth. To investigate the impact of political processes on interest rates, we continue to focus on parliamentary systems. Drawing on political science models of cabinet durability, we estimate the probability of a cabinet coming to an end – either due to a constitutionally mandated election or a loss of parliamentary support. Using the generated probability of a cabinet end as a proxy for market expectations, we show that the possibility of a cabinet coming to an end increases the costs of borrowing in both the public and private sectors.

The impact of information about political events on markets will vary according to whether market actors anticipate the event. In Chapters 6–8, we analyze how prior beliefs and expectations condition the influence of political events on market behavior. To do this, we focus on specific episodes of the political process: cabinet formations and elections. This allows us to estimate how the arrival of political information at particular times affects market behavior.

In Chapter 6, we examine bond market reactions to the process of cabinet negotiations. Cabinet negotiations are media events. Journalists cover the process closely, reporting on which parties are negotiating, what portfolios and policies are on the table, and what the likely outcome will be. To measure the dynamic process of cabinet bargaining, we perform content analysis of major newspapers during two instances of coalition bargaining: the surprising People's Party (ÖVP)–Freedom Party (FPO) coalition in Austria (1999–2000) and the National–New Zealand First coalition in New Zealand (1996). Using a Bayesian transformation of the information generated by the content analysis, we generate a proxy for market beliefs about which coalition is likely to form. We then test whether the bargaining process affects daily changes in government bond returns. We show that the impact of information about the bargaining process is conditioned by the prior beliefs of market actors about which coalition will form. Without an understanding of how markets update and process political information, however, we would not have been able to assess the market reaction to the cabinet negotiation process accurately.

Chapter 7 uses the case of the 2000 U.S. Presidential election to explore how the arrival of political information shapes market behavior. This analysis tracks the overnight movements of two major futures prices, the NASDAQ 100 Index and the S&P 500 Index, and the U.S. dollar–Japanese Yen exchange rate on election night. To measure the arrival of political information, we first calculate the probability of a Gore electoral college victory using state public opinion polls. When CNN called states for either candidate, we update this probability, providing a minute-by-minute estimate of the probability of a Gore victory throughout the night. Statistical analysis indicates that this updated probability affects both the price and the volatility (conditional variance) of these assets, again illustrating how the beliefs of market actors about political outcomes influence market volatility. By examining only the night of the election, the experimental design of this chapter limits the influence of earnings reports, unemployment or inflation announcements, or changes in interest rates. As a result, we are able to isolate the effect of political expectations on market behavior.

In Chapter 8, we show how excessive exchange rate movements can hurt an incumbent government's approval ratings by looking at the relationship between government popularity and exchange rate movements in Britain between 1987 and 2001. To generate an accurate measure of the short-term political costs of exchange rate fluctuations, however, we must account for the possibility that exchange rate behavior may also be a product of changes in the government's public standing. Indeed, the analyses in Chapter 3 imply that where public opinion shocks make political outcomes less certain, exchange rate volatility will increase. To deal with this endogeneity issue, we estimate separate models of exchange rate behavior and government voting intention iteratively and recursively. At each iteration, we use estimates from each model to generate measures of exchange rate and public opinion shocks. We then employ these generated variables in the next iteration of the estimates, including measures of political shocks in the model of exchange rates and measures of exchange rate behavior in the model of voting intention. We are able, therefore, to measure the political costs (in terms of government popularity) of exchange rate shocks and the exchange rate consequences (in terms of exchange rate behavior) of political competition.

IMPLICATIONS FOR DEMOCRATIC POLITICS

The results of our analysis establish (1) political uncertainty can affect financial market performance, and (2) financial market turbulence can negatively affect the ability of politicians to retain office. With a better understanding of the mechanisms that link democratic politics and market behavior, more informed discussion of the impact of larger more integrated financial markets on democratic politics becomes possible.

In particular, the results suggest that the market and political reforms in the industrial democracies over the past twenty years may be designed to break the linkages between financial market behavior and democratic political outcomes. These reforms include changes in the regulation of financial markets, monetary reforms (central bank independence, exchange rate commitments, the single currency in the European Union), and political reforms designed to reduce political uncertainty (electoral reform, devolution).

With the growth of capital markets over the past twenty years, political uncertainty carries greater economic costs. High levels of democratic political uncertainty may simply have become too costly from an economic perspective. Consequently, owners of assets are likely to demand political institutions that will produce predictable political outcomes and reduce their exposure to politically-induced financial market volatility.

In turn, politicians have found themselves increasingly vulnerable to financial market fluctuations. Politicians, therefore, have incentives to supply institutional reforms that limit their susceptibility to the negative political consequences of financial market shocks and increase political predictability to prevent financial market volatility.

The intersection of interests between the demanders of institutional reform (i.e., asset owners) and the suppliers of reform (i.e., politicians) provides an important background condition for the patterns of reform.

¹The basic outline of the argument extends back to Marx. The “classic” contemporary versions of the argument are Lindblom (1977) and Block (1977). For more recent arguments along these lines, see Garrett (1995), Goodman and Pauly (1993), Strange (1996), and Andrews (1994).

²We do not examine issues related to the origin, transformation, and regulation of markets across countries and over time. The literature on financial development is voluminous: for example, Rajan and Zingales (2003), North (1981), Neal (1990), Bernstein (2004), and Smith (2003).

³See, among others, Bachman (1992), Bernhard and Leblang (2002), Blomberg and Hess (1996), Christodoulakis and Kalyvitis (1997), Eichengreen, Rose, and Wyplosz (1995), Freeman, Hays, and Stix (2000), Leblang and Bernhard (2000, 2001), Leblang and Mukherjee (2004), Lobo and Tufte (1998).

⁴For example, Bachman (1992), Blomberg and Hess (1997), Freeman, Hays, and Stix (2000).

⁵For example, Alesina and Rosenthal (1995), Alesina, Roubini, and Cohen (1993), Alt and Crystal (1983), Franzese (2002), Fiorina (1991), Hibbs (1987), and Leblang and Mukherjee (1994).

⁶To explain these latter results, political economists speculate that investors shift to stocks as a hedge against the potentially inflationary consequences of a left government.

⁷Recently, some scholars have argued that, given certain conditions, market actors can deviate from pure rational behavior. Models in the “behavioral finance” literature propose that market equilibria can be understood in terms of “sunspots” or with reference to “fads” or “herd” behavior. Most importantly, this literature identifies the importance of non-economic information in providing focal points to economic actors. From this perspective, the theoretical arguments and empirical evidence presented in this book are consistent with the behavioral finance approach to financial market behavior. See Thaler (1992), Barberis and Thaler (2002), and Shleifer (2000) for surveys.

⁸This view, it should be noted, reflects an evolution in thought regarding efficient markets. When introduced into the economic literature thirty five years ago an efficient market was defined as one which “adjusts rapidly to new information” (Fama, Jensen, and Roll 1969).

⁹Malkiel (2003) puts it nicely: “A capital market is said to be efficient if it fully and correctly reflects all relevant information in determining security prices. Formally, the market is said to be efficient with respect to some information set … if security prices would be unaffected by revealing that information to all participants. Moreover, efficiency with respect to an information set … implies that it is impossible to make economic profits by trading on the basis of [that information set].”

¹⁰Even if an investor pays attention only to price movements, she will respond to the changes in the behavior of other traders who shift their assets based on political information. Only a relatively few traders, therefore, need to be informed about political processes in order for the market to move in response to political events.

¹¹One way to conceptualize the concept of volatility is as a confidence interval surrounding price forecasts. Assuming efficient markets, today's asset price change is the best predictor of tomorrow's price change, but the confidence interval surrounding the forecast estimate may be wide – indicating high volatility – or small – indicating low volatility. This idea is discussed in Chapter 7.

¹²The United States abandoned capital controls in 1974 and continued to ease financial market restrictions to attract foreign investors, particularly during the Reagan years. The United Kingdom abandoned capital controls in 1979 and in 1986 opened the London Stock Market to foreign securities firms, a move referred to as “The Big Bang.” Later in the 1980s, both Japan and countries in the European Union removed capital controls and liberalized the trading of domestic and foreign assets. See Helleiner (1994) for details of the major changes in financial markets during this period.

---

© Cambridge University Press

Table of Contents

1. Introduction; 2. Democratic processes and political risk: evidence from foreign exchange markets; 3. When markets party: stocks, bonds, and cabinet formations; 4. The cross-national financial consequences of political predictability; 5. Cabinet dissolutions and interest rate behavior; 6. Bargaining and bonds: the process of coalition formation and the market for government debt in Austria and New Zealand; 7. Time, shares, and Florida: the 2000 Presidential Election and stock market volatility; 8. Polls and pounds: exchange rate behavior and public opinion in Britain; 9. Conclusion: political predictability and financial market behavior.

Editorial Reviews

"How and how far can asset holders turn news into predictions about politics? How far do shifts in asset prices constrain political choices? Can politicians anticipate and deal with market-induced change? The answers in this innovative book come from careful, rigorous analysis of a wide range of exchange, bond, and stock markets, covering disparate political processes in many countries. The research is at the highest standard of empirical and theoretical analysis in political economy and political science today." James E. Alt, Harvard University

"Though political economy has had a wide influence in many parts of economics, the study of asset markets has largely been an exception. This book fills that gap, with its insightful and wide-ranging analysis of the effect of politics on stock, bond, and foreign exchange markets. It will be an important and much-needed addition to the bookshelves of students of politics, political economy, and finance." Allan Drazen, University of Maryland

"Democratic Processes and Financial Markets is a genuinely interdisciplinary contribution of tremendous significance. Building on literatures in finance, economics, and political science, Bernhard and Leblang analyze how, in a variety of institutional and historical settings, financial traders process political information. They then produce precise estimates of the magnitudes of political risk premia and of related impacts of politics. Out of their book emerge new and profound insights about the costs of democracy in general and of political uncertainty in particular." John R. Freeman, University of Minnesota

"In Democratic Processes and Financial Markets, William Bernhard and David Leblang show how the prices of financial assets – stocks, bonds, currencies – respond to such political developments as elections, cabinet formations and dissolutions, and trends in other nations. The authors make excellent use of current theoretical and empirical work in economics and political science. Their integration of approaches from the two disciplines is exemplary, and will undoubtedly influence many scholars who want to explore the interplay of economic and political phenomena. This excellent book will be a staple of the growing literature in political economy, and a model of how scholars can integrate theory and empirical work in the analysis of political and economic affairs." Jeffry Frieden, Harvard University

"This path-breaking book builds on the insight that an important part of the general relationship between democracy and capitalism is the relationship between the public events mandated by democratic political institutions and the behavior of asset markets. Using modern theories of finance, economics, and political science, this book provides a rich, deep and wide-ranging analysis that will be of great interest to anyone who wants to understand fully the relationship between these dominant political and economic systems." William R. Keech, Carnegie Mellon University

From the Publisher