Banque Nationale De Belgique Sa · EURONEXT:BNB
Banque Nationale De Belgique Sa — National Bank of Belgium — National Bank of Belgium Colloquium 2018: Understanding inflation dynamics: the role of co
21 slides from Banque Nationale De Belgique Sa's conference presentation, 2018
Slide 1: Markup and Price Dynamics: linking micro to macro
Markup and Price Dynamics: linking micro to macro Jan De Loecker1 Catherine Fuss2 Jo Van Biesebroeck1 1 KU Leuven and CEPR 2 National Bank of Belgium National Bank of Belgium Understanding inflation dynamics: the role of costs, mark-ups and expectations
Slide 2: Recent attention on micro-based aggregate markups
• Recent attention micro-based aggregate markups in US, and other regions – e.g. De Loecker and Eeckhout (2017, 2018). • Debate at various high-level policy circles (ECB Sintra, Jackson Hole, etc.) about the rise (or not) of markups and therefore market power. • Potentially large implications for st
Slide 3: This paper
This paper • This project: Belgium 1978-2016 • Small-open economy contrast - long panel 1978 onward. • Puts us in a unexplored territory: • Potential technological change (beyond Hicks-neutral): rise fixed cost? • Globalization: 1. competitive pressure, market size → substantive, 2. global firms (MN
Slide 4: Measurement of markups
Measurement of markups • We follow De Loecker and Warzynski (2011) and De Loecker and Eeckhout (2017) and apply the production-approach. • With data on firms (i) over time (t) on sales (Sit ), and a variable (V ) input’s expenditure (EitV ) , and an estimate of the corresponding output elasticity: S
Slide 5: What do we know?
What do we know? • Recent evidence on the US: rising markups and profits. • Listed Belgian firms (Worldscope): similar pattern.
Slide 6: Sample Construction
Sample Construction • Objective: construct long time-series of aggregate markup (and underlying determinants: shares and firm-level markups). • Challenge: data sources annual accounts (NBB AA): 78-84 and 85-16 • Non-trivial construction of sample capturing large share of Belgian private economic act
Slide 7: Implementation
Implementation • Crucial input to the method is the choice of a variable input of production: • Intermediate inputs: disqualifies capital and also labor (case of Belgium) • Cost of goods sold (not observed as such?), materials. • The latter greatly impacts pattern and interpretation: variables vs fi
Slide 8: Normalized Aggregate Markups: all intermediates
Normalized Aggregate Markups: all intermediates
Slide 9: Break down intermediate inputs
Break down intermediate inputs • Goods Intermediates (variable): raw materials used in production • Service intermediates (quasi-fixed): 1. Insurances 2. Transportation/Travel/Catering 3. Deliveries to the firm 4. Availability fees 5. Rent 6. Maintenances and repairs 7. Temporary and external work 8
Slide 10: Technology and Firm Organization
Technology and Firm Organization g g s yit = βst mit + βst L mits +βst K lit + βst kit + it | {z } Intermediates
Slide 11: Variable vs Fixed inputs
Variable vs Fixed inputs • First important observation: pattern of goods-intermediates and services-intermediates radically different.
Slide 12: Sectoral analysis
SECTORAL ANALYSIS Weighted mean 95th percentiles © © 4 = oO] ré 4 © o4 f~ = o - lo 4 + N J ~ a } 4 n = oI T T T T T T T ~ ~t T T T T T T T _ 1985 1990 1995 2000 2005 2010 2015 1985 1990 1995 2000 2005 2010 2015 Manufacturing (41% -> 32%) Trade (38% -> 43%) Agriculture (0.2%) Utilities (right axis, 7
Slide 13: Markup growth and reallocation
Markup growth and reallocation • Decompose ∆Mt into within and reallocation. • Consider two distinct decompositions: 1. Actual: āit = ait +a2 it−1 : X X X X āit ∆µit + µ̃it−1 ∆āit + µ̃it ait − µ̃it−1 ait−1 i∈I i∈I i∈En i∈Ex 2. Counterfactual: ait = ait−1 : X X X X X ait−1 ∆µit + µ̃it−1 ∆ait + ∆µi
Slide 14: Approach: top-to-bottom
Approach: top-to-bottom 1. Across/within sectors – aggregated across firms, • No entry and exit • Declining manufacturing sector (41 to 32 %) 2. Across/within firms –within a sector.
Slide 15: Decompositions: Across sectors
Decompositions: Across sectors • Period of markup growth due to markup growth within sectors, • Period of fluctuating markup: corr (∆ shares, ∆ markups)< 0.
Slide 16: Decompositions: Manufacturing
Decompositions: Manufacturing • Period of markup growth due to markup growth within firms, • Subsequent period growing firms lower margins (discuss US).
Slide 17: Decompositions: Trade
Decompositions: Trade • Period of markup growth due to markup growth within firms, • Trading firms exit had relatively high markups, • Subsequent period growing firms lower margins (discuss US).
Slide 18: Decompositions: take-aways
Decompositions: take-aways • Period of aggregate markup growth: within-firm markup growth. • Stagnating markup period: negative correlation between market share changes and markup growth. • Hypothetical decomposition suggests a potentially much more pronounced markup increase in absence of negative
Slide 19: Belgian competitiveness?
BELGIAN COMPETITIVENESS? Belgian imports Belgian exports Ln z4 bo L 1 4 + Eo = bw 2 cS 84 T T T T T 1 T T T T 1 T T 1 1988 1993 1998 2003 2008 2013 2018 988 1993 1998 2003 2008 2013 2018 Chinese import share ——-— E.Europe + N.Africa imp. share Total imports (right axis) Belgian share of total import
Slide 20: Major Findings (1): Technology and markups
Major Findings (1): Technology and markups • Fundamental change in production and sales process: rise fixed factors. • Increasing markups 1985-1995, in the overall economy, but a continuing rise up to 2005 in manufacturing. Remaining years stable. • Results are uniquely driven by the dynamics in the
Slide 21: Major Findings (2): Drivers
Major Findings (2): Drivers • Aggregate pattern masks dynamics at sector and firm level. • When markup rises, entirely due to the within markup growth. • When markup cycles around a stable average: strong process of reallocation (at the sectoral of firm level) → increasing negative correlation(∆ mar