Minjie Ding | Research

Research Line

How differences across households and countries reshape consumption behavior and monetary policy transmission.

Research Line

Where standard theory fails to account for the data, necessitating the development of new mechanisms.

MPC heterogeneity among households with asset threshold

Optimal Consumption under Asset Thresholds: Evidence and a Threshold HANK Model

Unlike standard theory, Chinese households’ propensity to consume does not decline monotonically with wealth; instead, it falls sharply in the early stage of wealth accumulation.
Using household survey data and DID/DDD designs, I identify the core mechanism as a deep bundling of household utility with housing thresholds, and I develop a Threshold HANK model to draw policy implications.
Conventional consumption policy would subsidize those with the strongest propensity to consume; I argue instead that more attention should be paid to households whose consumption has been suppressed by social mechanisms. As reliance on land finance recedes, we should loosen the link between household utility and housing—for example, by equalizing public-service rights for renters and homeowners. Counterfactual experiments show that such reforms can unlock substantial consumption without fiscal stimulus.
Explanations of weak monetary transmission often stress the supply side, arguing that credit is overly allocated to firms; I focus on the demand side, emphasizing that because many life entitlements are tied to housing, even when credit reaches households, some families use it for aggressive saving for future house rather than consumption.

MPC heterogeneity among food delivery riders

I joined Meituan as a food-delivery rider since December 2025, and collected consumption data on frontline food-delivery riders through offline interviews and online surveys. I obtained demographic, financial, discount-factor, and propensity-to-consume data from more than 8,000 riders.
The consumption behavior of China’s hand-to-mouth, as represented by food-delivery riders, differs markedly from standard economic theory. In particular, riders with lower discount factors (less patient) and with less wealth also have a low propensity to consume in practice, rather than the higher propensity that theory would predict.
Survey evidence shows that having a saving target is the core reason for this anomaly. Unmarried riders have the lowest propensity to consume because they are saving for marriage. Married riders at the same wealth level, by contrast, have a higher propensity to consume, consistent with theory.
Consumption policy typically targets subsidies at those with the highest propensity to consume, but groups who suppress consumption in order to save also deserve attention. Counterfactual experiments show that weakening the link between marriage and saving can unlock substantial consumption potential without fiscal subsidies.

Housing Market Heterogeneity

People usually look only at city-wide house-price trends, but homes differ in size, location, and property rights, and prices of different types of housing can move in different ways. I collected nearly 3 million existing-home transactions from Lianjia and Anjuke across more than 40 cities to study these heterogeneous price trends.
Using local projections, I study how U.S. monetary policy shocks affect house prices in China. After a U.S. tightening shock, prices fall more in cities with larger economies and a higher share of finance. Larger homes also see faster price declines, and their time to sale shortens markedly.

Country Heterogeneity

How do U.S. monetary tightening shocks affect other countries? The conventional view emphasizes the interest-rate channel: other countries follow with rate hikes, and output falls. Signaling theory instead argues that U.S. tightening conveys news of a stronger U.S. economy, which would raise output abroad.
Using VAR models and data for nearly 150 countries from 1995 to 2025, I find that in financially underdeveloped countries, the signaling view holds: both interest rates and output rise significantly. Inequality also shapes the output response: more unequal countries exhibit a weaker response.
The existing literature mainly studies U.S. monetary spillovers under different exchange-rate regimes. After controlling for exchange-rate regime, I further examine financial development, equality indices, and related factors.

Balassa-Samuelson Puzzle in China

This paper uses the Balassa-Samuelson framework to explain inflation differences across Chinese provinces, but finds that the relationship does not hold stably. Provinces with faster productivity growth in tradables do not necessarily see faster growth in the overall price level.
In theory, tradable prices should be similar across provinces. In China, however, large cross-province differences in land rents and labor costs also drive differences in tradable prices. After controlling for these distribution costs, the Balassa-Samuelson framework holds across provinces.
This study incorporates distribution costs into tradable prices and offers a perspective for refining the Balassa-Samuelson framework.