1

Reform prioritization and strategies: Complementarities, packaging, sequencing

Romain Duval Senior Advisor to the Chief Economist and Head of Structural Reforms Unit

Research Department International Monetary Fund

Structural Reforms 2.0 Workshop, France Strategie, Paris, January 12, 2018

Key reform interactions: holistic lessons from the literature

Non-linearities • Role of distance to frontier. Matters for prioritization and sequencing

Across reforms • High-dimensionality problem in 2nd-best world stay humble • Unclear whether there is reform complementarity across the board (e.g. labor and product market reforms) • But multiple specific interactions seem to matter and can be exploited Examples: UB-ALMPs, TW-min wage, PMR-capital market, R&D-tertiary education…etc • Political economy case for packaging may be stronger than economic case (past reformers)

Between reforms and business conditions • Tension between economics and political economy of reforms in good and bad times (e.g. labor market reforms)

Between reforms and macro policies • Macro policy support to reforms can ease the tension—and has done so in some success cases • Aggressive counter-cyclical macro policy may also support longer-term growth in deregulated product markets • ZLB and fixed exchange rate constraints: not serious issues—reforms are typically not/little deflationary 2

Some implications for reform strategies in EA context

Objectives and constraints • Objectives: amplify ongoing recovery; resilience to future shocks; LT growth and convergence • Constraints: Scarce fiscal space; scarce political capital; income inequality

Implications for prioritization/sequencing/packaging • Premium on reforms that pay-off quickly and create momentum—PMR, some LM reforms in good times… •…and reforms that rebuild fiscal space—aforementioned indirectly + public spending/sector reforms directly… • …that, in turn, can: i) facilitate adoption of other reforms, creating further momentum ii) enhance resilience to future shocks—directly and indirectly by creating macro policy space iii) down the road fund fiscally costly reforms that are key to longer-term growth (higher education, innovation) • Vast majority of reforms can be designed or packaged in ways that preserve or even enhance equity • EU/EA-level reforms in parallel, to enhance macro effect of domestic ones—over and above their own impact on resilience Example: capital market union + PMR reforms intangible investment and growth.

3

Thank You!

4

9.6

9.7

9.8

9.9

10

10.1

10.2

10.3

10.4

9.7

9.8

9.9

10

10.1

10.2

10.3

10.4

1984 1986 1988 1990 1992 1994 1996 1998

Australia synthetic

9.4

9.5

9.6

9.7

9.8

9.9

10

10.1

10.2

10.3

9.5

9.6

9.7

9.8

9.9

10

10.1

10.2

10.3

1984 1986 1988 1990 1992 1994 1996 1998

Netherlands synthetic

Case studies of major past reform packages (log GDP per capita, synthetic counterfactual obtained through synthetic control method)

Notes: Log(GDP per capita is shown). Estimates of counterfactual rely on the synthetic control method. Assumed treatment date = start of episode: AUS 1994, NZL 1991, NLD 1994. The two countries with the largest weights in the synthetic control groups are the United States and Greece (for New Zealand and Australia); Belgium and the United States (for the Netherlands). The number and estimated weights of other countries in the synthetic control groups vary across the four case studies. Vertical lines indicate the starting year of the reform episode. Source: Adhikari, Duval, Hu and Loungani (2016).

6% impact 1.5% impact

Case studies of major past reform packages (log GDP per capita, synthetic counterfactual obtained through synthetic control method)

Notes: Log(GDP per capita is shown). Estimates of counterfactual rely on the synthetic control method. Assumed treatment date = start of episode: AUS 1994, NZL 1991, NLD 1994. The two countries with the largest weights in the synthetic control groups are the United States and Greece (for New Zealand and Australia); Belgium and the United States (for the Netherlands). The number and estimated weights of other countries in the synthetic control groups vary across the four case studies. Vertical lines indicate the starting year of the reform episode. Source: Adhikari, Duval, Hu and Loungani (2016).

9.2

9.3

9.4

9.5

9.6

9.7

9.8

9.9

10

9.3

9.4

9.5

9.6

9.7

9.8

9.9

10

1981 1983 1985 1987 1989 1991 1993 1995

New Zealand synthetic

Effect of major past EPL reforms on sectoral employment: Unconditional impact vs. impact in major recessions (%)

7 Note: t=0 is the year of the major reform; dotted lines denote 90 percent confidence bands. The solid black (blue) line represents the unconditional effect (the effect in a major

recession). The chart shows a sectoral effect, measured as the differential effect of employment legislation reform for a sector with relatively high natural layoff rate constraints compared to a sector with a relatively low natural layoff rate. Source: Duval, Furceri and Jalles (2017).

Unconditional (average) effect

Effect in a major recession

Simulated impact of PMR and EPL reforms in normal times, bad times, and bad times with binding ZLB

-0,5

0,0

0,5

1,0

1,5

2,0

Year 1 Year 2 Long Run

Panel 1. GDP (Percent)

1. Effect of PMR reforms

2. Effect of EPL reforms

Panel 2. Unemployment (Percent)

Panel 3. GDP (Percent) Panel 4. Unemployment (Percent)

-1,0

-0,5

0,0

0,5

1,0

Year 1 Year 2 Long Run

-0,6-0,5-0,4-0,3-0,2-0,10,00,10,2

Year 1 Year 2 Long Run-0,2

0,0

0,2

0,4

0,6

0,8

Year 1 Year 2 Long Run

-5,00,05,0

Y… L…Normal times Bad times, unconstrained monetary policy Bad times, constrained monetary policy

8 Source: IMF (2016) based on Cacciatore, Duval, Fiori and Ghironi (2017).

Effect of major past product market reform packages on output and public debt (%)

9 Note: t=0 is the year of the major reform. Solid lines denote estimated effects. Dotted lines denote 90 percent confidence bands. Source: Banerji, Crispolti, Dabla-Norris, Duval, Ebeke, Furceri, Komatsuzaki and Poghosyan (2017).

Panel 1. Output Panel 2. Public Debt-to-GDP

-1

-0,5

0

0,5

1

1,5

2

2,5

3

3,5

4

-1 0 1 2 3 4 5 6

Perc

ent

Years

-10

-8

-6

-4

-2

0

2

4

-1 0 1 2 3 4 5 6

Perc

ent

of

GD

P

Years