Mallooki

Working paper / urban economics / August 2026

A Fuel Reform During a Moving Land-Price Gradient

Job access and registered property transactions in Kuwait

Kuwait's September 2016 fuel-price reform provides a clear commuting-cost mechanism but no untreated city. Using 23,033 registered transfers and fixed 2011 workplace access, the paper finds that a transaction-weighted peripheral land gradient flattens around the reform. The pattern is not geographically stable: Sabah Al-Ahmad Al-Marine lies 4.02 exposure standard deviations into the low-access tail and supplies 39.3 percent of effective land weight. Omitting it reverses every land parameter, while equal neighborhood-quarter weighting largely removes the pattern. The result is a case-driven spatial transaction cycle, not a stable causal fuel-capitalization effect.

Download PDFAll research
Evidence tier
Controlled descriptive
Data
23,033 transfers, 2013Q3 to 2019Q3
Outcome
Registered price per recorded plot m² among observed sales
Source
Kuwait Ministry of Justice; author's calculations

Question and mechanism

Higher gasoline prices can increase the value of job access in a car-oriented city. The paper asks whether registered property-price gradients by fixed pre-reform access changed around Kuwait's national fuel-price increase.

Design

Neighborhood exposure is predicted access to the 2011 workplace distribution, fixed before the reform. The analysis traces pre-slope, first-full-quarter break, and post-slope changes for recorded-land and completed-home transfers. It reports transaction weights, equal neighborhood-quarter weights, common geography, rolling pseudo-breaks, alternative access definitions, and leave-one-neighborhood-out estimates.

The outcome is log nominal registered transfer price per recorded plot square metre among observed sales. For completed homes, the numerator includes the house and plot and the denominator is plot area, not floor area. "Recorded land" is an administrative source category, not verified vacant land.

What the paper finds

Full-sample land post slope-0.001095% interval: -0.0143 to +0.0122 log points per quarter per SD lower access
Influential market39.3%Share of effective land weight from Sabah Al-Ahmad Al-Marine
Exposure tail4.02 SDSabah Al-Ahmad Al-Marine; next land neighborhood is 0.46 SD
Common-sample equality after omissionp = 0.789No stable land-versus-home path difference after removing the influential market

Source: Kuwait Ministry of Justice; author's calculations.

Two-panel scatterplot showing Sabah Al-Ahmad Al-Marine isolated at the low-access tail and carrying about 39 percent of effective recorded-land transaction weight, versus about 3 percent for completed homes.
One peripheral market is unusual in both exposure and land-market weight. This leverage—not a broad causal gradient—is the paper's central empirical result. Source: Kuwait Ministry of Justice; author's calculations.

The baseline gradient was already moving before the reform. Dropping Sabah Al-Ahmad Al-Marine flips every land parameter, and equal neighborhood-quarter weighting makes the land path nearly flat. The policy-centered local break is also not distinctive relative to nearby dependent pseudo-break dates.

What the paper does and does not claim

The study documents a spatial transaction cycle organized around a meaningful reform date and fixed access measure. It does not identify a causal reform effect, a property-stock price response, realized commuting behavior, household welfare, or a stable structural difference between administrative land and completed-home categories.

The next empirical step is to explain the development history, plot releases, infrastructure delivery, and sale incidence of Sabah Al-Ahmad Al-Marine in a balanced neighborhood-quarter panel that includes zero-sale cells.

Citation and data source

Suggested citation: Mallooki. 2026. "A Fuel Reform During a Moving Land-Price Gradient: Job Access and Registered Property Transactions in Kuwait." Working paper.

Transaction source: Kuwait Ministry of Justice, Live Registered Real-Estate Sale-Price Query. All reported estimates are the author's calculations from registered transactions. The public PDF contains aggregate estimates and no transaction rows, deed identifiers, or exact transaction locations.