4.3 - Monopolistic competition, TP uncertainty & Brexit

Are certain TP realistic

Firms may not know what tariffs in place in future but have to make trade decisions today

Under functioning international systems it can be realistic - WTO pre Trump

• Under WTO rules nations cant raise tariffs above bound rate

• Some uncertainty within bounds but mitigates a bit

• MFN tariffs dont move much in reality

PTAs more certain than general WTO MFN

Classic trade models

Ricardian model

• argues trade takes place due to CA

• CA caused by different productivities between sectors & nations

HO model

• Different factor abundance between nations + goods with different factor intensity drive trade

New trade model - Krugman 1979

Explain trade between similar countries (similar in productivity & factor endowment) - NOT EXPLAINED BY CLASSICAL MODELS

• Increasing RtS {Larger market size

• Consumer love for variety. {

• Monopolistic competition - each firm monopolist of residual D for its variety

Monopolistic comp with heterogenous firms - Assumptions

Basic model assumes homogenous firms - in reality no true

• Exporters more productive on avg BUT even among Exporters there is heterogeneity

More productive firms will charge lower p - to increase market share

Under heterogeneity, firms have to pay a sunk/fixed cost to enter a market - source of RtS

• Sunk/fixed - if paid before/after learning market conditions to enter

LR equilibrium of heterogenous firms model

Firms enter market until marginal firm equates Expected profits to sunk entry cost

E(pi) = sunk cost

Mono comp with heterogenous firms - graph

Firms have an idea of future profits

Firms only join the market if E(pi) > sunk cost

Only most productive join (productivity UP = pi UP)

2 doesnt enter but 1 does

Mono comp with heterogenous firms - with FT effects

Similar result to case with homogenous firms

Less productive dom firms exit

More productive F firms join & start exporting to H

Access to variety UP → Welfare UP

Productivity increases as least productive leave → P falls

Mono comp with heterogenous firms - graph

Assume no dom firms

H charges tariff on M - t a % of sales → profits fall

• Some F exporters wont enter market (pi < sunk)

• Consumer access to variety falls → CS down

Firms only enter if productivity > B

Firms with productivity A-B leave market

• firms reevaluate if they want to invest in X capital

TP uncertainty theory

Firms have to make investment decisions when entering a market - often under uncertainty

They decide based on Pi given information available today

Entry cost sunk upon joining BUT changing conditions can reduce Pi

• may force them to leave market & lose sunk cost

• markup p < initial investment

TP uncertainty example OV

Assume an event makes tariff rate uncertain - scenarios equally likely

2 firms - 2 is less productive than firm 1

• dont discount future profits

To X to F, each firm has to invest in X capital today

• Sunk cost = £23000 in this example

4 scenarios - tariffs rise in 2020,21,22 or never

• tariff UP = profits fall

TP uncertainty example RESULTS

If firms certain tariffs wont rise then both X

• 1: 30-23 > 0. 2: 24-23>0

However there is 75% chance tariffs rise

• 1: 27-23 > 0. 2: 21-23<0

• only 1 X if it decides to enter market

TP uncertainty graph

Firms with productivity B-C leave / dont join market - even though tariffs still t0

• uncertainty of rise forces them out

TP uncertainty effects

Fewer firms join market / more leave from the tariff uncertainty

Aggregate F X fall as no. exporters fall

• Varieties fall → CS falls

F firms that exit due to TP uncertainty could be replaces by less efficient dom firms

• Productivity down → P rise

• Therefore TPU works as protection for dom firms

Brexit uncertainty

2015 general election winners promised referendum

• created TP uncertainty

Caused EU & UK firms to lower E(pi) due to expected tariff rise

• decreased market entry

• CS fell Varieties down + M price index UP

Brexit uncertainty measured - Graziano et al (2020) OV

Used betting market probabilities as prob of Brexit + opinion polls

• showed uncertainty varied over time

Brexit uncertainty measured - Graziano et al (2020) RESULTS - impact on EU X to UK

Impact on EU X to UK relative to non-EU

• X fell sharply when referendum date announced

• X fell again when article 50 put in place (but recovered)

• Shows extreme uncertainty

Brexit uncertainty measured - Graziano et al (2020) RESULTS - Tariff threat & Trade

Paper estimated avg trade across affected products

• -11-20% trade between UK & EU

• Bilateral trade down -15% if MFN applied to all products

Brexit uncertainty measured - Graziano et al (2020) RESULTS - Uncertainty & M price

Import price captured changes in CS effectively

M price of EU goods relative to all M tracked closely with Brexit probabilities

• strong supportive evidence of model

If only focused on high risk of tariff products (proxy of UK tariff to EU was EU MFN tariffs)

• still tracks Brexit prob well

With mean tariff threat - M price index 12% increase

• Brexit uncertainty alone increased consumer prices by +0.5-0.7%