Measuring Economic Policy Uncertainty - Booth School of ...faculty.chicagobooth.edu/steven.davis/pdf/EconomicPolicy... · Measuring Economic . Policy Uncertainty . Steven J. Davis,

Measuring Economic Policy Uncertainty

Steven J. Davis, Chicago & NBER

Research with Scott Baker (joining Kellogg) And Nick Bloom (Stanford & NBER)

UC-San Diego April 2014

Uncertainty barometer

Economic Uncertainty: Multi-faceted and hard to measure


Economic Uncertainty: Multi-faceted and hard to measure But important in many theories & mechanisms: -- Option value of waiting -- Precautionary savings -- Cost of finance investment -- Ambiguity aversion -- Managerial risk aversion


Todays Focus: Newspaper-based measures of Economic Policy Uncertainty


Todays Focus: Newspaper-based measures of Economic Policy Uncertainty Consider other EPU measures too, partly as a cross-check on newspaper- based measures

Research Objectives 1. Construct new measures of Economic Policy

Uncertainty (EPU). Thus far: Monthly indexes for USA, Canada, UK, Germany,

France, Italy, Spain, China, India and Japan. Daily measure for USA

2. Develop methods to evaluate and refine the newspaper-based indexes.

3. Assess the view that policy-related economic uncertainty is at historically high levels in recent years in the U.S. and in Europe.

Which events and policy matters account for large changes in policy-related economic uncertainty?

6

Research Objectives 4. Assess the effects of policy uncertainty on firm-

level and macroeconomic performance As an impulse or driving force In amplifying and propagating other shocks (e.g.,

the financial crisis) 5. Related work: Understand economic, political

and social forces behind policy uncertainty Why Has U.S. Policy Uncertainty Risen Since

1960? BBD, Brandice Canes-Wrone and Jonathan Rodden, AER Papers & Proceedings (May 2014)

6. Underway: Surveys of subjective business expectations & perceived uncertainty Bloom & Davis w/ Atlanta Fed researchers & others

7

What Do We Want our EPU Index to Capture?

All of the following: Uncertainty about who will make economic policy

decisions e.g., who will win the next elections? Uncertainty about what economic policy actions

decision makers will undertake, and when. Uncertainty about the economic effects of policy

actions past, present and future actions Economic uncertainty induced by policy inaction Economic uncertainty related to national security

concerns and other policy matters that are not mainly economic in character

8

Constructing Our Newspaper-Based EPU Index for the U.S.

Search digital archives of 10 major newspapers for articles with terms related to EPU

For each newspaper: Get monthly EPU article counts Scale by count of all articles in same month Normalize scaled count so that SD=1 for 1985-2010 Newspaper-level EPU Index

Sum across newspaper-level indexes by month to get the U.S. EPU index

9


Text String Search Criteria: EU: {economic OR economy} AND {uncertain OR uncertainty} EPU: AND {regulation OR deficit OR federal reserve OR congress OR legislation OR white house}

10


Text String Search Criteria: EU: {economic OR economy} AND {uncertain OR uncertainty} EPU: AND {regulation OR deficit OR federal reserve OR congress OR legislation OR white house}

11

Later: How do we select the policy terms for our EPU search filter?


Newspapers: Boston Globe Chicago Tribune Dallas Morning News Los Angeles Times Miami Herald

New York Times SF Chronicle USA Today Wall Street Journal Washington Post

12

Note: We use Access World News Newsbank Service when constructing a daily EPU Index, because the daily index requires a higher density of news sources.

US Newspaper-Based Economic Policy Uncertainty (EPU) Index

Source: Measuring Economic Policy Uncertainty by Scott R. Baker, Nicholas Bloom and Steven J. Davis, all data at www.policyuncertainty.com. Data normalized to 100 prior to 2010.

50

100

150

200

250

Gulf War I

9/11

Clinton Election

Gulf War II

Bush Election

Stimulus Debate

Lehman and TARP

Euro Crisis and 2010 Midterms

Russian Crisis/LTCM

Debt Ceiling; Euro Debt

Black Monday

Fisc

al C

liff

Shu

tdow

n &

DC

2

Obama Election

January 1985 to February 2014

http://www.policyuncertainty.com/

European Newspaper-Based EPU Index 50

10

0 15

0 20

0 Po

licy

Unc

erta

inty

Inde

x

9/11

Asian Crisis

Papandreou calls for

referendum, then resigns

Italy Rating

Cut

German Elections

Lehman Bros.

Greek Bailout Request, Rating

Cuts

Northern Rock & Ensuing Financial Turmoil

Treaty of Accession/ Gulf War II

Nice Treaty Referendum

Source: www.policyuncertainty.com. Based on 10 paper (El Pais, El Mundo, Corriere della Sera, La Repubblica, Le Monde, Le Figaro, the Financial Times, Times, Handelsblatt, FAZ.)

Russian Crisis/LTCM

French and Dutch Voters Reject European Constitution

Ongoing Eurozone Stresses

January 1997 to February 2014


Two Measurement Concerns

Suitability: Whether an accurate count for news articles about a particular type of uncertainty provides a good indicator for that type of uncertainty.

Accuracy: Whether specific text-string search criteria accurately identify the set of articles that discuss a certain type of uncertainty, e.g., policy-related economic uncertainty.

15

Assessing Suitability Concern Idea: Apply news-based approach to a concept of uncertainty for which we have external, market-based evidence. Implementation: Compare VIX measure of uncertainty about future equity returns to a news-based index of equity market uncertainty, with search terms as follows:

{economic OR economy} AND {uncertain OR uncertainty} AND {stock price OR equity price OR stock market}

16

0 20

0 40

0 60

0 Figure 7: Newspaper-based index of equity market uncertainty compared

to market-based VIX, January 1990 to December 2012

Notes: The news-based index of equity market uncertainty is based on the count of articles that reference economy or economic, and uncertain or uncertainty and one of stock price, equity price, or stock market in 10 major U.S. newspapers, scaled by the number of articles in each month and paper. The news-based index and the VIX are normalized to a mean of 100 over sthe period.

Correlation=0.733 Monthly

News-Based Index of

Equity Market Uncertainty

Monthly Average of Daily VIX

Values

Political Slant in Newspaper Coverage of

Economic Policy Uncertainty

18

Figure 9: Political slant plays little role in time-series behavior of news-based EPU index

Source: Papers sorted into 5 most Republican and 5 most Democratic groups using the media slant measure from Gentzkow and Shapiro (2010).

0 10

0 20

0 30

0 40

0

1985 1990 1995 2000 2005 2010

Reagan, Bush I Bush II Clinton

Obama

5 most Republican papers

5 most Democrat papers

20

Assessing Accuracy Concern Working under our direction, RA teams read several thousand newspaper articles, following a 50-page audit guide to code articles: EU=0/1, EPU=0/1, and much more

Figure 8: Human Readings and Computer-Automated Method Yield Similar Newspaper-Based EPU Indexes, 1985Q1 to 2012Q2

Note: Based on random samples of 45 articles per quarter (fewer prior to 1993). For these articles, we calculate quarterly EPU rates based on human readings (red line) and based on computer-automated methods (blue line).

Using the Audit Sample and our preferred policy term set, correlation is 0.84 between human and computer methods.

Computer

Human

Other Audit Results Correlation of news-based EPU error rate and

real GDP growth rate = -0.12 (raw quarterly data) Correlation of EPU error rate and true EPU:

= -0.11 in the raw quarterly data = -0.06 in detrended quarterly data

Among EPU=1 articles: 5% are mainly about low or declining uncertainty 73% discuss uncertainty about what or when 32% discuss uncertainty about effects 15% discuss uncertainty about who

22

Other Audit Results

23

Uncertainty about Who

Who Uncertainty Times EPU Value

All Years 15% of EPU Articles

16

Presidential Election Years

37% of EPU Articles

42

What/When Uncertainty

Times EPU Value

Effects Uncertainty

Times EPU Value All Years 77 34

Presidential Election Years

70 28

Selecting a Preferred Term Set Consider 16,000 permutations of 14 policy-

relevant terms: regulation, budget, spending, policy, deficit, tax, federal reserve, white house, house of representatives, government, congress, senate, president, and legislation.

+ 14,000 combinations that replace terms like policy and government with multi-word terms like government policy

Interpreting the human coding as truth, select the term set that minimizes the sum of false positive and false negative error rates

24

Error rates for 28,000 permutations of 14 policy terms in a human audit sample of 3,500 randomly selected articles

Permutations of 14 policy terms: regulation, budget, deficit, tax, federal reserve, government, congress, senate, president, legislation, government spending, federal spending, etc.

False positives and false negatives expressed as a fraction of true EPU as judged by human readers.

Our preferred policy term set: {congress, deficit, federal reserve, legislation, regulation, white house}

More Sophisticated Methods? Our EPU measure is simple and transparent:

Based on presence of certain terms, classify each newspaper article as EPU = 1 or 0.

Automate and search over vast digital archives. Count the 1s, scale, and aggregate over papers.

Can we build a better EPU index using more sophisticated classification methods in machine-learning literature? We are investigating.

Practical constraints: Archives dont reside on our disk Search platforms have limited capabilities 26

Three Other Types of Indicators of Economic Policy Uncertainty for the United States

Tax Code Expirations Index Discounted revenue impacts of federal tax code

provisions set to expire over next 10 years Based on Congressional Budget Office projections

Frequency counts for uncertainty and policy-related uncertainty in the Federal Reserves Beige Books: Beige Book issued every six weeks in advance of each regularly

scheduled FOMC meeting since 1983 Survey-based measures of disagreement among professional

forecasters about future inflation and future government purchases of goods and services. See paper or back of slide deck.

27

Figure 3: Federal Tax Code Expirations Index, 1991-2013

Notes: Based on Congressional Budget Office data on projected revenue effects of federal tax code provisions set to expire in the current calendar year and next ten years. For a given year, the index value is calculated as the discounted sum of projected revenue effects associated with expiring tax code provisions, using a discount factor of 0.5^T applied to future revenue effects for T=0,1,10 years. Index normalized to a mean of 100 before 2010.

Fede

ral T

ax C

ode

Expi

ratio

n In

dex

0 50

0 10

00

1500

Figure 3: Federal Tax Code Expirations Index, 1991-2013

Notes: Based on Congressional Budget Office data on projected revenue effects of federal tax code provisions set to expire in the current calendar year and next ten years. For a given year, the index value is calculated as the discounted sum of projected revenue effects associated with expiring tax code provisions, using a discount factor of 0.5^T applied to future revenue effects for T=0,1,10 years. Index normalized to a mean of 100 before 2010.

Fede

ral T

ax C

ode

Expi

ratio

n In

dex

0 50

0 10

00

1500

Undiscounted projected 10-year revenue impact of scheduled tax code expirations: Before 2003 < $250 billion 2009-2012: $3-5 trillion

2013: Huge drop due to Fiscal Cliff resolution

30

Counts of Uncertainty and Policy-Related Uncertainty per Beige Book, 1983Q3-2013Q4

Source: Baker, Bloom and Davis review of all FOMC Beige Books, available at www.federalreserve.gov/monetarypolicy/fomc_historical.htm.

0

10

20

30

40

1983

Q3

1984

Q3

1985

Q3

1986

Q3

1987

Q3

1988

Q3

1989

Q3

1990

Q3

1991

Q3

1992

Q3

1993

Q3

1994

Q3

1995

Q3

1996

Q3

1997

Q3

1998

Q3

1999

Q3

2000

Q3

2001

Q3

2002

Q3

2003

Q3

2004

Q3

2005

Q3

2006

Q3

2007

Q3

2008

Q3

2009

Q3

2010

Q3

2011

Q3

2012

Q3

2013

Q3

Average Uncertainty CountAverage Policy-Related Count

Tax Cut Expiration Battle

Debt- Ceiling Fight

Iraq Invasion

Fiscal Cliff & Sequester

Shu

tdow

n &

DC

2

http://www.federalreserve.gov/monetarypolicy/fomc_historical.htm

Evidence about Sources of Policy Uncertainty

31

Note: Analysis uses Newsbank coverage of around 1000 US national and local newspapers See Table 1 in the Baker, Bloom and Davis (2013) for a more detailed analysis.

What drives recent rise in U.S. economic policy uncertainty?

PresenterPresentation NotesSmall newspapers like the Modesto Bee or the Oregon Emerald, Seattle Times

Note: Analysis uses Newsbank coverage of around 1000 US national and local newspapers See Table 1 in the Baker, Bloom and Davis (2013) for a more detailed analysis.

What drives recent rise in U.S. economic policy uncertainty? Newspaper articles point mainly to concerns about fiscal and healthcare policies

PresenterPresentation NotesSmall newspapers like the Modesto Bee or the Oregon Emerald, Seattle Times

Beige Books Tell a Similar Story Counts By Category and Selected Time Period, 1983Q3 to 2013Q1

1990 Q4 - 1991 Q1

Gulf War I

1993 Q2 - 1993 Q3

Clinton Tax Reforms

2001 Q4 - 2002 Q2

9/11 Attacks

2002 Q4 - 2003 Q2

Gulf War II

2004 Q2 - 2004 Q4

Bush/Kerry Election

2008 Q3 - 2009 Q4

Lehman's and recession

2010 Q1 - 2013 Q1

Debt-ceiling crisis

1983 Q3 2013 Q1 Overall Average

Overall Economic Uncertainty 11 8.8 7.7 13.5 5.2 10.2 15.8 5.5 Economic Policy Uncertainty 5.5 6.3 1.2 4.8 2.8 0.8 6.8 1.7 All Fiscal Matters 1 5.5 1.5 0 0 0.4 3.3 1.0

Taxes Only 0 3.3 0.2 0 0 0.3 1.4 0.4 Spending Only 0.5 1 1 0 0 0.2 1.2 0.3

Monetary Policy 0 0 0 0 0 0 0 0 Health Care 0 2 0 0 0 0.2 0.5 0.1 National Security and War 5.3 0.3 0 2 0 0 0.1 0.2 Financial Regulation 0 0 0 0 0 0.2 1.2 0.2 Sovereign debt, currency crisis 0 0 0 0 0 0 0.8 0.1 U.S. Elections and Leadership Changes 0 0 0 0.2 2.2 0 0.9 0.2 Other Specified Policy Matters 0 0.5 0.7 0 0.2 0 0.5 0.2 Politics, Unspecified 0.5 1 0 3 0.7 0 1.6 0.3 Sum of Policy & Politics Categories 6.8 9.3 2.2 5.2 3.0 0.8 10.0 2.5

But also Highlight Concerns Related to Financial Regulation, Sovereign Debt, and Politics

1990 Q4 - 1991 Q1

Gulf War I

1993 Q2 - 1993 Q3

Clinton Tax Reforms

2001 Q4 - 2002 Q2

9/11 Attacks

2002 Q4 - 2003 Q2

Gulf War II

2004 Q2 - 2004 Q4

Bush/Kerry Election

2008 Q3 - 2009 Q4


2010 Q1 - 2013 Q1

Debt-ceiling crisis





And NEVER Breathe a Word about about Monetary Policy Uncertainty!

1990 Q4 - 1991 Q1 Gulf

War I

1993 Q2 - 1993 Q3

Clinton Tax Reforms

2001 Q4 - 2002 Q2

9/11 Attacks

2002 Q4 - 2003 Q2

Gulf War II

2004 Q2 - 2004 Q4

Bush/Kerry Election

2008 Q3 - 2009 Q4


2010 Q1 - 2013 Q1

Debt-ceiling crisis





Big 5 Sources of Economic Policy Uncertainty, 1985Q1 to 2013Q3

0

20

40

60

80

100

120

140

160

Jan-

85Ju

l-85

Jan-

86Ju

l-86

Jan-

87Ju

l-87

Jan-

88Ju

l-88

Jan-

89Ju

l-89

Jan-

90Ju

l-90

Jan-

91Ju

l-91

Jan-

92Ju

l-92

Jan-

93Ju

l-93

Jan-

94Ju

l-94

Jan-

95Ju

l-95

Jan-

96Ju

l-96

Jan-

97Ju

l-97

Jan-

98Ju

l-98

Jan-

99Ju

l-99

Jan-

00Ju

l-00

Jan-

01Ju

l-01

Jan-

02Ju

l-02

Jan-

03Ju

l-03

Jan-

04Ju

l-04

Jan-

05Ju

l-05

Jan-

06Ju

l-06

Jan-

07Ju

l-07

Jan-

08Ju

l-08

Jan-

09Ju

l-09

Jan-

10Ju

l-10

Jan-

11Ju

l-11

Jan-

12Ju

l-12

Jan-

13Ju

l-13

Monetary Fiscal Health Security Regulation

Note: This chart is a quarterly version of Table 1 in Measuring Economic Policy Uncertainty by Baker, Bloom and Davis. It shows the 5 most important sources of economic policy uncertainty based on frequency counts of newspaper articles.

U.S. Economic Policy Uncertainty: A Longer Term Perspective

38

Notes: Index of Policy-Related Economic Uncertainty composed of quarterly news articles containing uncertain or uncertainty, economic or economy, and policy relevant terms (scaled by the smoothed total number of articles) in 5 newspapers (WP, BG, LAT, WSJ and CHT). Data normalized to 100 from 1900-2011.

Newspaper-Based EPU Index from 1900 Scaled by Economy/ic articles rather than all articles January 1900 December 2012

100

200

300

1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000 2010

Polic

y U

ncer

tain

ty In

dex

Versailles conference

9/11 and Gulf War II

Debt Ceiling

OPEC II

Lehman and TARP Great

Depression, New Deal and FDR

Post-War Strikes, Truman-Dewey

Great Depression Relapse Gulf

War I Black Monday

Start of WW I

OPEC I

Russian Crisis

Watergate

Assassination of McKinley

Gold Standard Act

Berlin Conference

McNary Haughen farm bill

PresenterPresentation NotesPolitical fighting:Taft v Roosevelt over Conservatives vs Progressives in Republican Party (1910-1913)The November 1912 spike is the election of Woodrow Wilson over Roosevelt (second) and Taft (third)

Close elections:Wilson v Hughes in 1916Truman vs Dewey in 1948JFK-Nixon 1960

Major economics policies:Woodrow Wilsons New Freedom progressive agenda (1913-1921)Calvin Coolidge (1923-29) very laissez-faire and inactive (no new experiments), reduced size of government and debt, and scientific taxation (minimize tax changes see http://en.wikipedia.org/wiki/Calvin_Coolidge)New Deal 1932 plus last year under HooverJFK New Frontier starts in 1960LBJ Great Society from 1964 (like New Frontier uses task forces of outsiders, including lots of economists)

Large daily stock market moves, 1900-2012

Correlation of number of policy-triggered jumps per year with EPU index is 0.78

Policy increases Non-Policy increases

Policy decreases Non-Policy decreases

Reproduced from What Triggers Large Stock Market Jumps? by Baker, Bloom & Davis

Based on human readings of next-day news articles About large S&P Index moves in the New York Times And the Wall Street Journal. Jump threshold: +/- 2.5%

Assessing the Effects of EPU: Summary of our Work

41

1.Micro approach: Exploit differences in exposure to government spending to estimate the effects of EPU on firm-level implied volatility, investment rates and employment growth

1.Macro approach: Include our EPU measure in otherwise standard VAR models of aggregate dynamics. Estimate the effects of EPU innovations on aggregate output, investment and employment.

Microdata: Our firm-level panel regressions exploit differences in industry exposure to government spending

Source: Authors calculations using the Federal Registry of Contracts (1999-2013) matched to Compustat firms (using Compustat parent & D&B names).

Note: We assign an intensity value of 0.43 to Health Services based on government spending share in this sector.

Microdata Results: Firms in sectors with greater exposure to government spending have higher stock volatility when EPU is high

Magnitude of Implied Volatility Effects Summary Statistics Mean firm-level implied volatility = 0.5 Time-Series st. dev. of firm-level implied vol.: -- 0.14 for low-exposure firms -- 0.13 for high-exposure firms

Illustrative Calculation Consider a 100 log-point increase in EPU for a firm with

an exposure level of 0.25. Use IV results to get implied impact on firms implied

volatility: (0.25)(1.0)(1.17)=0.29 -- About 60% of mean firm-level volatility -- More than twice the average time-series st. dev.

45

Using annual firm-level data from 2000.

Microdata Results: High EPU depresses investment And employment growth at firms in sectors with high exposure to government spending.

46

For a firm with average investment rate and average contract exposure, doubling EPU results in estimated investment fall of only 0.08 percentage points (~1.2% fall).

For firms in the 90th percentile of exposure rates, the impact is much larger, with predicted investment drops of 0.8-5.0 percentage points, depending on specification and baseline investment.

Magnitude of Firm-Level Investment Effects

Figure 12: Estimated Industrial Production and Employment after a Policy Uncertainty Shock

Indu

stria

l Pro

duct

ion

Impa

ct

(% d

evia

tion)

Months after the economic policy uncertainty shock

Notes: This shows the impulse response function for Industrial Production and employment to an 102 unit increase in the policy-related uncertainty index, the increase from 2006 (the year before the current crisis) to 2011. The central (black) solid line is the mean estimate while the dashed (red) outer lines are the one-standard-error bands. Estimated using a monthly Cholesky Vector Auto Regression (VAR) on the EPU index, log(S&P 500 index), federal reserve funds rate, log employment, log industrial production and linear time trend. Fit to data from 1985 to 2011.

Empl

oym

ent I

mpa

ct

(mill

ions

)

-5-4

-3-2

-10

12

0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36

-3-2

-10

0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36

-4-3

-2-1

0Es

timat

ed im

pact

on

indu

stria

l pro

duct

ion

0 5 10 15 20 25 30 35

Figure 13: Robustness of Estimates to Different VAR Specifications

Months after the policy uncertainty shock Notes: This shows the impulse response function for GDP and employment to an 102 unit increase in the policy-related uncertainty index. Estimated using a monthly Cholesky Vector Auto Regression (VAR) of the uncertainty index, log(S&P 500 index), federal reserve funds rate, log employment, log industrial production and time trend unless otherwise specified. Data from 1985 to 2011.

Indu

stria

l Pro

duct

ion

Impa

ct

(% d

evia

tion)

Baseline

Bivariate (uncertainty and log industrial production)

Three months of lags

Nine months of lags

Uncertainty index has equal weight

on measures Adding VIX and putting it first

Assessing the Effects of EPU: Selected Other Work

49

1.Geography of Great Recession: Shoag and Veuger (2013) find larger unemployment rate rises from 2006 to 2009 in states with (a) greater increases in state-level uncertainty and (b) institutions less well suited for mediating the effects of a general rise in uncertainty.

2. International Spillovers: The IMFs World Economic Outlook (2013) finds that increases in U.S. and European EPU reduce growth in other regions of the world, with bigger spillover effects from U.S. EPU.

3.Firm-Level Studies: Julian and Yook (2010) and Gulen and Ion (2013) find that political and policy uncertainty retard business investment. Durnev (2010) finds a lower sensitivity of investment to stock prices in election years.

50

Methodology: New methods to construct, evaluate and refine measures of economic uncertainty based on frequency counts of newspaper articles and other text sources.

Factual Claim: U.S. EPU levels are at historically high levels from 2008-2013. Major episode in the 1930s as well.

Data Products: New monthly indices of EPU for U.S., China, Germany, Japan, Spain, France, Italy, U.K., India, and Canada. A new daily EPU index for the United States

Summary

51

Evidence of EPU Effects: Firm-level regressions: high levels of EPU

reduce investment and employment growth at firms with high exposure to government contract awards, and raise stock volatility

Simple VAR models: increases in EPU foreshadow declines in output, investment and employment.

Correlates of Rising EPU: Secular growth in U.S. EPU coincide with growth of government and increasing political polarization Our paper with Canes-Wrone & Rodden

Summary

More on Economic Policy Uncertainty at www.policyuncertainty.com


End of Slides for Prepared Remarks

53

PresenterPresentation NotesWhy So Much Political Gridlock and Such High Policy Uncertainty?

References Bernanke, Ben, 1983. Irreversibility, Uncertainty and Cyclical Investment, Quarterly Journal of Economics,

98, pp. 85106. Carroll, Royce, Jeff Lewis, James Lo, Nolan McCarty, Keith Poole, and Howard Rosenthal, 2008. Who Is

More Liberal, Senator Obama or Senator Clinton? 18 April. Durnev, Art, 2010. The Real Effects of Political Uncertainty: Elections and Investment Sensitivity to Stock

Prices, working paper, McGill University, September. Fernandez-Villaverde, Jesus, Guerron-Quintana, Pablo, Kuester, Keith and Juan Rubio-Ramirez, 2011.

Fiscal volatility shocks and economic activity, University of Pennsylvania mimeo. Gentzkow, Matthew and Jesse Shapiro, 2010. What Drives Media Slant? Evidence from U.S. Daily

Newspapers, Econometrica, 78, 35-72. Gilchrist, Simon, Jae W. Sim and Egon Zakrajsek, 2010. Uncertainty, Financial Friction and Investment

Dynamics, working paper. Gulen, Huseyin and Mihai Ion, 2013. Policy Uncertainty and Corporate Investment, Working Paper,

Purdue. International Monetary Fund, 2013. World Economic Outlook: Hopes, Realities, Risks, April. Julio, Brandon and Youngsung Yook, 2010. Political Uncertainty and Corporate Investment Cycles, Journal

of Finance, forthcoming. Narita, Futoshi, 2011. Hidden Actions, Risk-Taking, and Uncertainty Shocks, University of Minnesota,

February. Panousi, Vasia and Papanikolaou, Dimitris, 2012. Investment, idiosyncratic risk and ownership, Journal of

Finance Pastor, Lubos and Veronesi, Pietro, 2012. Uncertainty about government policy and stock prices, Journal

of Finance, 67, 1219-1264. Pastor, Lubos and Veronesi, Pietro, 2013. Political Uncertainty and Risk Premia, forthcoming, Journal of

Financial Economics. Shoag, Daniel and Stan Veuger, 2013. Uncertainty and the Geography of the Great Recession, AEI

Economics Working Paper 2013-05, September. 54

Two Broad Classes of Explanations:

1) Expanding Scale and Scope of Government

2) Increasing Political Polarization These two types of explanations are not mutually exclusive, and they could overlap. Other explanations are also possible.

What Drives the Secular Rise in EPU Since 1960?

050

100

150

200

250

New

s ec

onom

ic p

olic

y un

certa

inty

1950 1960 1970 1980 1990 2000 2010

US Newspaper-based policy uncertainty index shows a strong upwards trend since the 1960s

Source: www.policyuncertainty.com Data normalized to 100 across the whole period. Papers covered are Boston Globe, Chicago Tribune, LA Times, New York Times, Wall Street Journal and Washington Post.

JFK assassination

9/11

OPEC I

Black Monday

Lehman and TARP

Debt Ceiling; Euro Debt

OPEC II

Fiscal Cliff and

Shutdow

n

Volcker Rate rises Gulf

War I

Gulf War II


2025

3035

40Go

vern

men

t sha

re o

f GDP

050

100

150

200

250

Polic

y un

certa

inty

1950 1960 1970 1980 1990 2000 2010year

One source of rising policy uncertainty could be the expansion of Government spending

Notes: Total government spending as a % of GDP from the US BEA (federal, state, and local spending on consumption, investment and transfer payments). See http://www.bea.gov/faq/index.cfm?faq_id=552

Polic

y U

ncer

tain

ty

Gov

ernm

ent s

pend

ing,

% o

f GD

P

http://www.bea.gov/faq/index.cfm?faq_id=552

Notes: Page counts from the Code of Federal Regulations (CFR), which describes all Federal regulations in effect in a given year, CFR data from Dawson and Seater (2013), spliced to data from Crews (2013) for 2006 to 2012. The CFR and Policy Uncertainty series are scaled to 100 from 1949 to 2012.

0

50

100

150

200

250

Polic

y U

ncer

tain

ty

Cod

e of

Fed

eral

Reg

ulat

ions

Pag

e C

ount

Another source of rising policy uncertainty could be the increasing volume of Government regulation

PresenterPresentation Notes1

Notes: Page counts from the Code of Federal Regulations (CFR), which describes all Federal regulations in effect in a given year, CFR data from Dawson and Seater (2013), spliced to data from Crews (2013) for 2006 to 2012. The CFR and Policy Uncertainty series are scaled to 100 from 1949 to 2012.

0

50

100

150

200

250The size and complexity of the federal tax code has also increased enormously in recent decades.

Polic

y U

ncer

tain

ty

Cod

e of

Fed

eral

Reg

ulat

ions

Pag

e C

ount

Another source of rising policy uncertainty could be the increasing volume of Government regulation

PresenterPresentation Notes1

Republican-Democrat differences in Congressional voting patterns point to rising political polarization

Notes: DW-Nominate index is the difference between the mean Republican and mean Democratic DW-Nominate roll-call ideal points generated by McCarty, Poole, and Rosenthal. Bonica (2013) has created ideological scores for each member of Congress based on scaling of campaign contribution records. We display the difference between the means of Republicans and Democrats in Congress.

The electorate also appears to perceive the two parties as increasingly polarized

Notes: DW-Nominate index is the standard-deviation of the first dimension of the DW-Nominate roll-call idealized points. The percentage seeing large difference between Democrats and Republicans in the American National Electoral Survey.

In contrast, U.S. voter preferences are unimodal and show little evidence of polarization

Notes: This scale was created by Ansolabehere, Rodden, and Snyder (2006) from responses to a variety of questions on economic policy issues appearing in each General Social Survey from 1974 to 2004.

1) Implied volatility of equity returns (e.g., VIX)

2) Newspaper-based measures (e.g., our EPU index)

3) Forecaster disagreement & uncertainty (e.g., Philly Fed)

4) Qualitative surveys (e.g., our Beige Book count)

5) Output volatility (e.g., GDP GARCH)

6) Dispersion and volatility of business-level outcomes

= Available in real-time

= Available with a delay of several weeks or so

Economic Uncertainty Measures & Proxies

Forecaster Disagreement about Future Inflation and

Future Government Purchases

66

.5

1 1.

5

Gulf War II & Fed Interest Rate Cuts

Gulf War I

Clinton Election

Budget Enforcement

Act

Obama Election, Banking

Crisis

Balanced Budget Act

Figure 4: Interquartile Range of Four-Quarter-Ahead CPI Inflation Forecasts, Percentage-Point Spread, Q1 1985 to Q4 2012

Notes: From the Federal Reserve Bank of Philadelphia Survey of Professional Forecasters (made every quarter; offset one month due to release dates such that Q4 covers Nov-Jan. Displays the Interquartile (IQ) range of the quarterly 1-year-ahead forecasts of CPI.

IQ R

ange

of A

nnua

lized

Infla

tion

Rat

e Fo

reca

sts

QE2 and Fed Statements

0 .0

05

.01

.015

.0

2 .0

25

IQR

of G

over

nmen

t Pur

chas

es F

orec

asts

, % o

f GD

P

Notes: Based on data from the Federal Reserve Bank of Philadelphia Survey of Professional Forecasters. We compute the interquartile range (IQR) of 1-year ahead forecasts of government purchases of goods and services and scale the IQR by the median forecast. We carry out these calculations separately for federal purchases and state & local purchases, then aggregate using the purchases share of nominal GDP for each level of government. See the main text for additional details.

Figure 5: Interquartile Range of Government Purchases Forecasts, Q1 1985 Q2 2013

Balanced Budget Act

Clinton Election

9/11

Budget Enforcement Act

Obama Election, Banking

Crisis

Debt Ceiling Dispute

Gulf War II

Economic Policy Uncertainty Indexes for Other Countries

69

China EPU Index, Jan1995-Jun2013, Based on South China Morning Post

70

0

50

100

150

200

250-New Prime Minister (Kaifu) -Structural Impediments Initiative (March 2-4)

Gulf War Soviet August Coup

- G7 Summit in Tokyo - Election (LDP less than 1/2 seats in the House)

NATO Summit (Brussel)

G7 Summit in Scotia, canada (June 15-17)

Prime Minister switched back to LDP

Collapse of Yamaichi Securities (1 of 4 Major Securities Companies)

New Prime Minister (Obuchi)

House of Representatives Dissolve & General Election Prime Minister

(Mori) announces stepping down

Prime Minister Koizumi reforms his cabinet and Heizo Takenaka to be the minister of Economic, Fiscal and Financial Policy to deal with nonperforming loans (NPLs)

Iraq War House of Representatives Disolution postponed

Nikkei drops to 8567

BOJ intervenes the exchange rate market to sell yen for dollars (1st time in 6.5 yrs)

Democratic Party announces possible tax increase

- State of Union Address by President Bush - FRB lowers Fed Funds Rate by 0.75%

House of Representatives Dissolve

Japan EPU Index, Sep. 1983 to June 2013, Based on Yomiuri and Asahi

71

50

100

150

200

250

2003 2008 2013

72

India Policy Uncertainty Index January 2003 to June 2013

Using a 50% weight on six major Indian newspapers and a 50% weight on forecaster disagreement measures. Constructed in collaboration with Sanjai Bhagat, Pulak Ghosh and Srivivasan Rangan. Downloaded from www.PolicyUncertainty.com on 7 July 2013

Congress Party wins National

Election

Bear Sterns

India-US Nuclear Deal

Lehman Failure Price

Hikes

Exchange Rate Concerns

Lokpal Bill

http://www.PolicyUncertainty.com

0 10

0 20

0 30

0

1997 2002 2007 2012

UK Economic Policy Uncertainty Index

Notes: UK Policy Uncertainty Index composed of news component and dispersion measures regarding budget balance and consumer prices forecasts. News measure composed of the monthly number of news articles containing uncertain or uncertainty, economic or economy, as well as policy relevant terms (scaled by the smoothed number of articles containing the word today). Policy relevant terms include: policy, tax, spending, regulation, Bank of England, budget, and deficit. Index covers January 1997 to June 2013. Papers include the Financial Times and the Times of London. Forecast dispersion component data from Consensus Economics forecasts.

9/11 Russian Crisis

Papandreou call for referendum

European Constitution Rejection/Summit

Lehman Bros.

Greek Bailout Request,

Rating Cuts

Northern Rock Takeover

Treaty of Accession/ 2nd Gulf War

European Constitution

Signing

Spanish Bank

Bailout Debates

50

100

150

200

250

300

1997 2002 2007 2012

9/11

France Economic Policy Uncertainty Index

Russian Crisis

Papandreou calls for referendum,

later resigns


Lehman Bros.

Greek Bailout Request, Rating Cuts

Northern Rock Support

Treaty of Accession/ 2nd Gulf War

Nice Treaty Referendum

Fran

ce P

olic

y U

ncer

tain

ty In

dex

European Sovereign Debt

Crisis

Notes: France Economic Policy Uncertainty Index composed of news component and dispersion measures regarding budget balance and consumer prices forecasts. News measure composed of the monthly number of news articles containing uncertain or uncertainty, economic or economy, as well as policy relevant terms (scaled by the smoothed number of articles containing today). Policy relevant terms include: policy, tax, spending, regulation, central bank, budget, and deficit. Index covers Jan 1997 June 2013. Papers include the Le Monde and Le Figaro. Searches done in native language of country. Forecast dispersion component data from Consensus Economics forecasts.

50

100

150

200

250

1997 2002 2007 2012

9/11

Germany Economic Policy Uncertainty Index

Russian Crisis

Papandreou call for

referendum; later resigns

Italy Rating Cut

German Elections

Eurozone Stresses


Rating Cuts

Northern Rock Support/ Takeover

Treaty of Accession And Gulf War

Ger

man

y Po

licy

Unc

erta

inty

Inde

x

Notes: Germany Economic Policy Uncertainty Index composed of news component and dispersion measures regarding budget balance and consumer prices forecasts. News measure composed of the monthly number of news articles containing uncertain or uncertainty, economic or economy, as well as policy relevant terms (scaled by the smoothed number of articles containing today). Policy relevant terms include: policy, tax, spending, regulation, central bank, budget, and deficit. Index covers Jan 1997 Jun 2013. Papers include the FAZ and Handelsblatt. Searches done in native language of country. Forecast dispersion component data from Consensus Economics forecasts.

Lehman Bros.

50

100

150

200

250

1997 2002 2007 2012

9/11

Italy Economic Policy Uncertainty Index

Asian Crisis

Papandreou call for referendum;

later resigns Italy Rating Cut


Lehman Bros.


Rating Cuts Treaty of

Accession Gulf War II

Italy

Pol

icy

Unc

erta

inty

Inde

x

Notes: Italy Economic Policy Uncertainty Index composed of news component and dispersion measures regarding budget balance and consumer prices forecasts. News measure composed of the monthly number of news articles containing uncertain or uncertainty, economic or economy, as well as policy relevant terms (scaled by the smoothed number of articles containing today). Policy relevant terms include: policy, tax, spending, regulation, central bank, budget, and deficit. Index covers Jan 1997 Jun 2013. Papers include the Corriere della Serra and La Repubblica. Searches done in native language of country. Forecast dispersion component data from Consensus Economics forecasts.

0 50

10

0 15

0 20

0 25

0

2001 2006 2011

Spai

n Po

licy

Unc

erta

inty

Inde

x

9/11

Spain Economic Policy Uncertainty Index

Lehman Bros. Greek Bailout

Request, Rating Cuts

Treaty of Accession And Gulf War II

Madrid Bombings

European Sovereign

Debt Stresses


Euro Adoption

Northern Rock Support/ Takeover

Notes: Spain Economic Policy Uncertainty Index composed of news component and dispersion measures regarding consumer prices forecasts. News measure composed of the monthly number of news articles containing uncertain or uncertainty, economic or economy, as well as policy relevant terms (scaled by the smoothed number of articles containing today). Policy relevant terms include: policy, tax, spending, regulation, central bank, budget, and deficit. Index covers Jan 2001 Jun 2013. Papers include El Mundo and El Pais. Searches done in native language of country. Forecast dispersion component data from Consensus Economics forecasts.

Canadian Economic Policy Uncertainty Index

Update

50

100

150

200

250

Can

adia

n Po

licy

Unc

erta

inty

Inde

x

Source: www.policyuncertainty.com. Created with help from Dorinda So from the Institute for Competitiveness & Prosperity www.competeprosper.ca

1st Gulf War

9/11

Clinton Election

2nd Gulf War

Debt Ceiling

Dispute; Euro Debt

Canadian Election and Lehman Bros.

Quebec Referendum

Canadian Elections

Canadian Election

http://www.policyuncertainty.com/http://www.competeprosper.ca/

More on the Audit and Another Suitability Check

79

Running the Newspaper Article Audit 1. Design, evaluate, and refine audit template 2. Define the Audit Universe: All articles coded

EU=1 by automated search 3. Sample Audit Universe and manually read articles

Randomly sample 3 articles per month for 5 of the newspapers; 45 articles per quarter

4. Code each article: EU, EPU, type of EPU, etc. 5. Compare manually read truth to results from

automated search with various permutations of policy terms

80

2 4

6 8

10

50

100

150

200

250

1950 1960 1970 1980 1990 2000 2010

Une

mpl

oym

ent n

ews

sear

ch

Une

mpl

oym

ent r

ate

Notes: Index of Unemployment News composed of quarterly news articles containing terms like unemployment, layoffs, or job loss (scaled by the smoothed total number of articles) in 5 newspapers (WP, BG, LAT, WSJ and CHT). Data normalized to 100 from Jan 1900-Dec 2011. Unemployment data is overall seasonally adjusted unemployment rate taken from the BLS.

Suitability check: news based indices for tracking unemployment also seem to work well

Correlation=0.72

More Checks and Comparisons

82

0 .0

1 .0

2 .0

3

1985 1988 1991 1994 1997 2000 2003 2006 2009 2012

Debt Ceiling Articles Government Shutdown Articles

Debt Ceiling and Government Shutdown in U.S. Newspapers, Jan. 1985 to Nov. 2013

Fraction of Articles In U.S. Newspapers

1995 Shutdown

Summer 2011 Debt-Ceiling Dispute

Fisc

al C

liff

This chart reports frequency counts of articles containing debt ceiling or government shutdown, expressed as a fraction of all articles in U.S. newspapers covered by Access World News Newsbank Service (1000+ newspapers).

Note: Data for November 2013 are based on part of the month.

Par

tial G

over

nmen

t Shu

tdow

n an

d D

ebt C

eilin

g 2

Somewhat Different Experiences in Great Depression and Great Recession

Notes: Index of Policy-Related Economic Uncertainty composed of quarterly news articles containing uncertain or uncertainty, economic or economy, and policy relevant terms (scaled by the smoothed total number of articles) in 5 newspapers (WP, BG, LAT, WSJ and CHT). Data normalized to 100 from 1900-2011.

50

100

150

200

250

300

New

s-Ba

sed

EPU

0 .0

1 .0

2 .0

3 .0

4 .0

5

Rea

lized

Sto

ck V

olat

ility

1928 1930 1932 1934 1936

Realized Stock Volatility News-Based EPU

50

100

150

200

250

300

New

s-Ba

sed

EPU

0 .0

1 .0

2 .0

3 .0

4 .0

5

Rea

lized

Sto

ck V

olat

ility

2006 2008 2010 2012

Lehman and

TARP Stimulus Debate

Debt Ceiling

Stock Crash

100 Days, New Deal, Gold Std.

2nd New Deal; SS;

NLRB Depression Bottom; Bonus

Army; Revenue Act

Comparison Newsbank (daily data summed to the monthly level) with our 10 paper series

Different weighting schemes for index components give similar results

50

100

150

200

Baseline (main) Economic Policy Uncertainty Index

Principal Component Factor Index

Equal-Weighted Index (1/4 each on z-scores)

50

100

150

200

250

300

1985 1990 1995 2000 2005 2010

Circulation weighting in news-based index component matters little C

ircul

atio

n W

eigh

ted

Equa

l Wei

ghtin

g

Notes: Index of Policy-Related Economic Uncertainty composed of quarterly news articles containing uncertain or uncertainty, economic or economy, and policy relevant terms (scaled by the smoothed total number of articles) in 5 newspapers (WP, BG, LAT, WSJ and CHT). Data normalized to 100 from Jan 1900-Dec 2011. Government expenditure is total federal, state, and local expenditures over GDP, annually.

10

20

30

40

50

60

VIX

(red

)

50

100

150

200

250

Econ

omic

Pol

icy

Unc

erta

inty

Inde

x (B

lue)

1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012

9/11 WorldCom & Enron

Gulf War II

Credit Crunch

Asian crisis

Gulf War I

Obama Election, Banking Crisis

Debt Ceiling

LTCM default

US index is similar to the VIX index of 1 month implied S&P500 stock market volatility, but not the same

Correlation VIX and Policy Uncertainty is 0.55

Large interest rate cuts

Clinton election

Source: www.policyuncertainty.com. Data until October 2012


Notes: Data from The buzz: Links between policy uncertainty and equity volatility, by Krag Gregory and Jose Rangel, Goldman Sachs, November 12, 2012.

Correlation EPU and 1 month=0.578 Correlation EPU and 10 years=0.855

10

20

30

40

50

60

50

100

150

200

250

Eco

nom

ic p

olic

y un

certa

inty

2002 2004 2006 2008 2010 2012

Implied volatility

1 Month Implied Volatility ()

10 Year Implied Volatility (+)

Economic Policy Uncertainty ()

EPU Index Is More similar to 10 year index of implied volatility on the S&P500 (correlation 0.73)

93

Key variable of interest is the interaction between aggregate EPU change and firm-level exposure to government contract awards.

Firm-level exposure measures are time invariant. We calculate exposure as weighted SIC-level government contract intensity, with weights given by firms own industry distribution of sales.

94

Controls include interactions of firm-level exposure with current change and forecasted future change in government purchases.

And interaction with change in VIX, plus firm and time fixed effects.

Economic Uncertainty: Multi-faceted and hard to measureEconomic Uncertainty: Multi-faceted and hard to measureBut important inmany theories &mechanisms:-- Option value of waiting-- Precautionary savings-- Cost of finance investment -- Ambiguity aversion-- Managerial risk aversionTodays Focus: Newspaper-based measures of EconomicPolicy UncertaintyTodays Focus: Newspaper-based measures of EconomicPolicy UncertaintyConsider other EPU measurestoo, partly asa cross-checkon newspaper-based measuresResearch ObjectivesResearch ObjectivesWhat Do We Want our EPU Index to Capture?Constructing Our Newspaper-Based EPU Index for the U.S.Constructing Our Newspaper-Based EPU Index for the U.S.Constructing Our Newspaper-Based EPU Index for the U.S.Constructing Our Newspaper-Based EPU Index for the U.S.Slide Number 13Slide Number 14Two Measurement ConcernsAssessing Suitability ConcernFigure 7: Newspaper-based index of equity market uncertainty compared to market-based VIX, January 1990 to December 2012 Political Slant in Newspaper Coverage of Economic Policy UncertaintyFigure 9: Political slant plays little role in time-series behavior of news-based EPU indexSlide Number 20Figure 8: Human Readings and Computer-Automated Method Yield Similar Newspaper-Based EPU Indexes, 1985Q1 to 2012Q2 Other Audit ResultsOther Audit ResultsSelecting a Preferred Term SetError rates for 28,000 permutations of 14 policy terms in a human audit sample of 3,500 randomly selected articlesMore Sophisticated Methods?Three Other Types of Indicators of Economic Policy Uncertainty for the United States Slide Number 28Slide Number 29Counts of Uncertainty and Policy-Related Uncertainty per Beige Book, 1983Q3-2013Q4 Evidence about Sources of Policy Uncertainty Slide Number 32Slide Number 33Beige Books Tell a Similar Story Counts By Category and Selected Time Period, 1983Q3 to 2013Q1 But also Highlight Concerns Related to Financial Regulation, Sovereign Debt, and PoliticsAnd NEVER Breathe a Word about about Monetary Policy Uncertainty! Big 5 Sources of Economic Policy Uncertainty,1985Q1 to 2013Q3U.S. Economic Policy Uncertainty: A Longer Term PerspectiveSlide Number 39Large daily stock market moves, 1900-2012Assessing the Effects of EPU: Summary of our WorkMicrodata: Our firm-level panel regressions exploit differences in industry exposure to government spendingMicrodata Results: Firms in sectors with greater exposure to government spending have higher stock volatility when EPU is highMagnitude of Implied Volatility EffectsSlide Number 45Magnitude of Firm-Level Investment Effects Slide Number 47Slide Number 48Assessing the Effects of EPU: Selected Other WorkSummarySummaryMore on Economic Policy Uncertaintyat www.policyuncertainty.comEnd of Slides for Prepared RemarksReferencesSlide Number 55Slide Number 56Slide Number 57Slide Number 58Slide Number 59Slide Number 60Republican-Democrat differences in Congressional voting patterns point to rising political polarizationThe electorate also appears to perceive the two parties as increasingly polarizedIn contrast, U.S. voter preferences are unimodal and show little evidence of polarizationEconomic Uncertainty Measures & ProxiesForecaster Disagreement about Future Inflation and Future Government PurchasesSlide Number 67Slide Number 68Economic Policy Uncertainty Indexes for Other CountriesChina EPU Index, Jan1995-Jun2013, Based on South China Morning Post Japan EPU Index, Sep. 1983 to June 2013, Based on Yomiuri and Asahi Slide Number 72UK Economic Policy Uncertainty IndexSlide Number 74Slide Number 75Slide Number 76Slide Number 77Slide Number 78More on the Audit andAnother Suitability CheckRunning the Newspaper Article AuditSlide Number 81More Checks and ComparisonsDebt Ceiling and Government Shutdown in U.S. Newspapers, Jan. 1985 to Nov. 2013Slide Number 84Comparison Newsbank (daily data summed to the monthly level) with our 10 paper seriesSlide Number 86Slide Number 87US index is similar to the VIX index of 1 month implied S&P500 stock market volatility, but not the sameSlide Number 89Slide Number 93Slide Number 94

Documents

Measuring Economic Policy Uncertainty - Booth School of ...faculty.chicagobooth.edu/steven.davis/pdf/EconomicPolicy... · Measuring Economic . Policy Uncertainty . Steven J. Davis,