The Price of Political Uncertainty: Theory and Evidence from the Option Market

• Published date: 01 October 2016
• Author: PIETRO VERONESI,BRYAN KELLY,ĽUBOŠ PÁSTOR
• Date: 01 October 2016
• DOI: http://doi.org/10.1111/jofi.12406

THE JOURNAL OF FINANCE •VOL. LXXI, NO. 5 •OCTOBER 2016

The Price of Political Uncertainty: Theory and

Evidence from the Option Market

BRYAN KELLY, ˇ

LUBO ˇ

SP´

ASTOR, and PIETRO VERONESI∗

ABSTRACT

We empirically analyze the pricing of political uncertainty, guided by a theoretical

model of government policy choice. To isolate political uncertainty, we exploit its vari-

ation around national elections and global summits. We findthat politicaluncertainty

is priced in the equity option market as predicted by theory.Options whose lives span

political events tend to be more expensive. Such options provide valuable protection

against the price, variance, and tail risks associated with political events. This pro-

tection is more valuable in a weaker economy and amid higher political uncertainty.

The effects of political uncertainty spill over across countries.

POLITICAL UNCERTAINTY HAS FEATURED PROMINENTLY in the economic landscape

of recent years. In the United States, much uncertainty surrounded the gov-

ernment bailouts during the financial crisis of 2008, the reforms of finance

and health care, the Federal Reserve’s innovative monetary policy, tax policy,

as well as the political brinkmanship over the debt ceiling in 2011 and 2013.

Standard & Poor’s named political uncertainty as a key reason for its first-

ever downgrade of the U.S. Treasury debt in 2011. In Europe, the sovereign

debt crisis has witnessed tremendous uncertainty about the actions of Euro-

pean politicians, central bankers, and Greek voters. Alas, despite the salience

of political uncertainty, our understanding of its effects on the economy and

financial markets is only beginning to emerge.

∗All authors are at the University of Chicago Booth School of Business and NBER. P´

astor

and Veronesi are also at the CEPR. P´

astor is also a member of the board of the National Bank

of Slovakia. The views in this paper are those of the authors, not of the institutions they are

affiliated with. We are grateful for comments from participants at the 2015 AFA meetings, 2015

NBER Summer Institute, 2015 NYU/Penn Conference on Law and Finance, 2015 Queen Mary

Conference on Recent Advances in Finance, 2014 Asset Pricing Retreat at Tilburg, 2014 Duke/UNC

Asset Pricing Conference, 2014 EFAmeetings, 2014 ESSFM Gerzensee, 2014 Princeton conference

on “The Causes and Consequences of Policy Uncertainty,” 2015 FARFE conference, 2014 Rising

Stars Conference, 2014 SoFiE Conference, and seminars at Bocconi, Chicago, EDHEC, Imperial,

Pennsylvania, Rochester, UCLA, Wisconsin, and the National Bank of Slovakia. We also thank

Jonathan Brogaard, Andrea Buraschi, Hui Chen, John Cochrane, George Constantinides, Ian Dew-

Becker, Michael Gofman, Brandon Julio, Juhani Linnainmaa, Rajnish Mehra, Toby Moskowitz,

Stavros Panageas, Jay Shanken, Ken Singleton, Raghu Sundaram, Eric Talley, Moto Yogo, and

two anonymous referees. Gerardo Manzo and Sarah Kervin provided excellent research assistance.

This research was funded in part by the Fama-Miller Center for Research in Finance and by the

Center for Research in Security Prices at the University of Chicago Booth School of Business.

DOI: 10.1111/jofi.12406

2417

2418 The Journal of Finance R

Recent research suggests that uncertainty about government actions has

negative real and financial effects. For example, Baker,Bloom, and Davis (2013)

find that this uncertainty,as measured by their index, increases unemployment

and reduces investment. Using the same index, P´

astor and Veronesi (2013)

find that this uncertainty commands a risk premium, and that stocks are

more volatile and more correlated in times of high uncertainty. Fern´

andez-

Villaverde et al. (2012) find that uncertainty about fiscal policy, as measured

by time-varying volatility of tax and spending processes, has negative effects

on economic activity.

A key obstacle in assessing the impact of political uncertainty is the difficulty

in isolating exogenous variation in this uncertainty.Political uncertainty likely

depends on a host of factors, such as macroeconomic uncertainty. The time-

series measures of Baker, Bloom, and Davis (2013) and Fern´

andez-Villaverde

et al. (2012) might potentially reflect not only government-related uncertainty

but also broader uncertainty about economic fundamentals.

In this paper, we isolate political uncertainty by exploiting its variation

around major political events, namely, national elections and global summits.

We investigate whether and how the uncertainty associated with these events

is priced in the option market. Options are uniquely well suited for this analy-

sis, for two reasons. First, they have relatively short maturities, which we can

choose to cover the dates of political events. An option whose life spans a politi-

cal event provides protection against the risk associated with that event. Since

the political event is often the main event that occurs during the option’s short

life, the option’s price is informative about the value of protection against po-

litical risk. Second, options come with different strike prices, which allow us to

examine various types of risk associated with political events, such as tail risk.

Elections and summits are also well suited for our analysis because they can

result in major policy shifts. Moreover, the dates of elections and summits are

determined sufficiently far in advance so that they are publicly known on the

dates at which we calculate the prices of the options spanning the events. These

events thus provide a source of exogenous variation in political uncertainty.

Our analysis is guided by the theoretical model of P´

astor and Veronesi (2013,

hereafter PV). In this model, the government decides which policy to adopt

and investors are uncertain about the future policy choice. We use this model

directly when analyzing global summits. When we analyze national elections,

we reinterpret the PV model so that voters decide whom to elect and investors

face uncertainty about the election outcome. Under the election interpretation,

political uncertainty is uncertainty about who will be elected; in the original

version of the model, it is uncertainty about which government policy will be

chosen.1

1Both interpretations of political uncertainty are mentioned in the following recent quote:

“When economists talk of political risk, they usually mean...national elections...But there is another

kind of political risk: the temptation for governments of all political colours to change the rules,

whether they relate to tax, the way that companies operate or how markets behave. And that risk

has increased significantly since the 2008 crisis” (The Economist, Buttonwood, November 9, 2013).

The Price of Political Uncertainty 2419

We derive the option pricing implications of the PV model, obtaining a closed-

form solution for the prices of put options whose lives span a political event.

We calculate three option market variables: the implied volatility of an at-

the-money (ATM) option, the slope of the function relating implied volatility

to moneyness, and the variance risk premium. These variables capture the

value of option protection against three aspects of risk associated with political

events: price risk, tail risk, and variance risk, respectively. In response to a

political event, stock prices might drop (price risk), the price drop might be

large (tail risk), and return volatility might rise (variance risk). These risks

are correlated inside the PV model; for example, the adoption of a risky policy

can raise volatility while depressing stock prices. The model implies that all

three option market variables should be larger, on average, than the same

variables calculated for options whose lives do not span a political event.

The model also predicts a negative relation between all three option market

variables and economic conditions. This prediction follows from a key result

in the PV model: when the economy is weaker, the current government policy

is less likely to be retained, creating uncertainty about which new policy will

be adopted instead. Similarly, in the election version of the model, the incum-

bent government is less likely to be reelected in a weaker economy, creating

uncertainty about the new government. Options provide protection against an

unfavorable policy decision or an undesirable election outcome. Since the prob-

ability of such an outcome is higher in a weaker economy, so is the value of the

option protection.

Finally, the model predicts that all three option market variables should be

larger amid higher political uncertainty.When there is no such uncertainty, op-

tion prices are governed by the Black-Scholes formula, so that implied volatility

equals expected volatility and both the implied volatility slope and the variance

risk premium are zero. When the uncertainty is present, the option protection

against all three aspects of political risk is valuable.

In our empirical analysis, we test the model’s predictions on option data

from 20 countries. We analyze two types of political events: national elections

and global summits. Each summit counts as a separate event for all countries

participating in the summit, while elections count only for the country in which

they are held. For each political event, we calculate the three option market

variables based on options whose lives span the event, adjusting each variable

for its mean calculated for neighboring options whose lives do not span the

event.

We find strong empirical support for the model’s predictions. First, the un-

conditional means of all three variables are significantly positive, for both elec-

tions and summits. The average (mean-adjusted) implied volatility is 1.43%

per year, which implies that one-month ATM put options whose lives span

political events tend to be 5.1% more expensive than neighboring options.2

2The 5.1% price premium is economically large. For example, if one were to purchase enough

one-month ATM put options to insure the total market value of the S&P 500 index in March 2014,

the additional cost for options that span political events would be over $15 billion. To obtain this