Platform/Industry & Policy Lab
Industrial policy and market analysis
Industry & Policy Lab

Industrial economics, built for practitioners.

From HHI and UPP to DID and DEA — every tool competition economists, regulators, and policy analysts need. Upload your data and get defensible results in minutes.

HHI · UPPDID · RDD · IVTFP · DEAMerger SimulationWelfare Analysis
35+
Analysis Methods
5
Sections
14
Categories
AI
Auto Interpretation

Markets and policy — answering with evidence?

01Can you say — with numbers — whether this market is competitive?
02Can you tell in advance whether a merger will harm competition?
03Can you catch collusion or abuse of dominance in the data?
04Can you prove the policy actually worked — causally?

From HHI to DID — Industry & Policy Lab answers with evidence.

01 · STRUCTURE

How is the market built?

Define markets, measure concentration, assess entry.

Identify the relevant market with SSNIP-based tests and cross-price elasticity. Quantify market power with HHI, Lerner Index, and markup estimation. Detect barriers to entry and track firm entry and exit rates.

SUMMARY

Unconcentrated, competitive market

HHI 1451 · 8 entities · top entity 23% share

Entities

8

HHI

1451

CR4

64%

Top entity

Cola_A

Concentration · Revenue share

Cola_A
Cola_B
Cola_C
Juice_A
Lemon_A
Water_A
Juice_B
Water_B

Cross-price elasticity (top 5)

Q↓ / P→ColaAColaBColaCJuiceALemonA
ColaA-2.21-0.73-0.76-1.32-1.66
ColaB-2.00-1.08-1.30-1.52-1.58
ColaC-1.67-0.80-1.64-1.68-3.41
JuiceA-0.05-1.19-2.38-1.91-1.27
LemonA-0.87-0.78-1.19-0.87-1.00

02 · CONDUCT

How do firms behave?

Detect collusion, dominance abuse, and vertical restraints.

Screen for price parallelism and bid-rigging patterns. Quantify abuse of dominance with predatory pricing and price discrimination tests. Assess vertical integration and foreclosure risk.

SUMMARY

Substantial pricing power detected

Mean Lerner 41.0% but only 67% of firms align with demand-implied benchmark

MEAN LERNER

41.0%

22 firms

DEMAND-IMPLIED

44.5%

1/|elasticity|

AGREEMENT RATE

67%

|est−theo| <10pp

CONDUCT THETA

1.02

0=Bertrand,1=Monopoly

Estimated vs demand-implied Lerner

Demand-Implied Lerner (%)Estimated (%)

03 · M&A

Will this merger harm competition?

Screen, simulate, and measure merger effects.

Apply ΔHHI, UPP, and diversion ratio tests to screen merger risk. Simulate post-merger prices with Logit-Bertrand. Measure stock market and price effects with event studies and difference-in-differences.

Scale Economies

Increasing returns to scale

η = 1.20 > 1: cost rises slower than output. Larger firms have a cost advantage.

α_Q = 0.834 (SE 0.013)

η = 1/α_Q = 1.199 → IRS

MES ≈ Q = 335 (AC = 2.18)

R² = 0.944

α_Q

0.834

SE 0.013

η=1/α_Q

1.199

IRS

MES (Q)

335

AC = 2.18

AC_max/AC_min

1.49

cost dispersion

Average Cost Curve (Mean AC by Q-bin)

MES0.851.72.553.4Output Q (log scale)

TOP GUPPI DIRECTIONS

Ranked by GUPPI = D_ij × (P_j − MC_j) / P_i

Solid bars exceed the 10% screening threshold

B-Cola→A-Cola
11.4%
B-Diet→A-Cola
10.2%
A-Cola→B-Cola
9.4%
A-Diet→B-Cola
9%
A-Lemon→B-Cola
7.4%
B-Cola→A-Diet
6.9%
B-Diet→A-Diet
6.5%
A-Cola→B-Diet
5.9%
A-Diet→B-Diet
5.6%
A-Lemon→B-Diet
3.9%
B-Cola→A-Lemon
3.2%
B-Diet→A-Lemon
3.1%

04 · POLICY

Did the policy work?

Causal identification — DID, RDD, IV, event study.

Estimate causal effects of regulations, subsidies, and trade policy using difference-in-differences, regression discontinuity, and instrumental variables. Measure tax pass-through and price regulation impact.

TAX / SUBSIDY PASS-THROUGH

How much of a cost change reaches price?

ρ = 1 is full pass-through; ρ < 1 means the firm absorbs part of the cost

ρ (level FE)

0.705

SE 0.018, t = 39.03

ρ (first-diff)

0.733

SE 0.057

Cumul. (0–3 lag)

0.651

long-run approx

Test ρ ≠ 1

Reject

vs full pass-through

Asymmetric Pass-Through

Separate ρ for cost increases and decreases — a gap signals "rockets and feathers"

Cost up (ρ⁺)

0.705

SE 0.082 · n = 558

Cost down (ρ⁻)

0.671

SE 0.298 · n = 132

ρ⁺ − ρ⁻ = 0.033 (SE 0.309, t = 0.11) · Symmetric (cannot reject)

Distributed-Lag Pass-Through

Heavy lag-0 means immediate response; spread-out coefficients mean gradual adjustment

0
1
2
3

PRICE REGULATION

Price distribution

Density histogram — the spike at the cap is the visible bunching mass

226507498122146169
PriceBunching at cap

05 · IMPACT

What are the economic effects?

Productivity, dynamics, linkages, and welfare.

Measure total factor productivity with Levinsohn-Petrin. Estimate scale economies and X-inefficiency with DEA. Track firm survival and reallocation. Quantify output and employment multipliers and welfare effects.

SCALE ECONOMIES

log AC vs log Q regression

OLS slope is α_Q; if α_Q < 1, the line slopes down (IRS)

024680.40.81.21.6log Qlog AC

CONSUMER & PRODUCER WELFARE

Market diagram & counterfactual price shifts

Demand and supply curves with equilibrium

P*QP

Counterfactual price shifts

Price shiftP_newQ_newCSPSDWL
+5%52.8094.90-244.2854.785.66
+10%55.3290.40-477.2398.2322.65
+20%60.3581.39-909.15151.1490.62

ALL IN ONE LAB

Everything Industry & Policy Lab does.

35+ methods across structure, conduct, M&A, policy, and impact.

Market definition

SSNIP-based tests and cross-price elasticity to draw the relevant market.

Market power

HHI & concentration, Lerner index, markup, labor monopsony.

Entry & exit

Entry barriers and firm entry/exit rates.

Collusion screening

Price parallelism and bid-rigging patterns.

Abuse of dominance

Predatory pricing and price-discrimination tests.

Vertical relations

Vertical integration and foreclosure risk.

Merger screening

ΔHHI, UPP, and diversion ratios.

Merger simulation

Logit-Bertrand post-merger price simulation.

Causal identification

DID, RDD, IV, and event studies for policy effects.

Regulation impact

Tax/subsidy pass-through, price regulation, trade policy.

Productivity & dynamics

TFP (Levinsohn-Petrin), scale economies (DEA), firm survival, reallocation.

Linkages & welfare

Output/employment multipliers, consumer & producer welfare, wage premium.

Defensible analysis, faster.

35+ methods. Upload your data and get publication-ready results in minutes.

No credit card required
35+ analysis methods