Multiple exchange rate systems have been adopted in many emerging markets to regulate the prices of essential goods and mitigate consumption inequality. In Iran, in the spring of 2022, the Tarjihi exchange rate for most essential goods was abolished, and cash transfers to households increased. These policy changes provide an opportunity to examine the effects of multiple exchange rate policies on consumption inequality. The present study examined the impact of multiple exchange rate policies on consumption inequality in Tehran, using daily sales data from one of the city’s major retail chains, Consumer Price Index (CPI) data, and household income data from the Refah Database. The analysis covered 16 of Tehran’s 22 districts, representing different income levels. It also employed a triple-difference-in-differences (DDD) approach to estimate the policy’s effects. The results showed that overall consumption inequality across Tehran’s districts did not change significantly. However, inequality in consumption of certain product categories, particularly meat and dairy products, increased in the medium term. This study is the first to examine the impact of the policy package that eliminated the Tarjihi exchange rate and increased cash transfers on consumption inequality. The findings provide valuable evidence for policymakers seeking to evaluate the effectiveness of multiple exchange rate policies and their implications for household consumption.
Introduction
In 2018, Iran introduced a subsidized exchange rate—known as the Tarjihi rate—of 4,200 tomans per US dollar for essential imports, substantially below the prevailing market rate. In May 2022, the government abolished this Tarjihi rate for most essential goods and increased monthly cash transfers for most households from approximately 95,000 to 300,000–400,000 tomans per person.
The distributional effects of this policy package are theoretically ambiguous. Eliminating the subsidized exchange rate raises the relative prices of affected goods, potentially placing a disproportionate burden on lower-income households, which typically allocate a larger share of their expenditure to essential goods and have fewer opportunities to substitute toward alternatives. However, the simultaneous increase in cash transfers may partially offset these losses.
The present study aimed to analyze the effects of multiple exchange rate policies on consumption inequality in Iran. It sought to address the following research questions: (1) Did the removal of the Tarjihi exchange rate, combined with the increase in cash transfers, affect consumption inequality across Tehran’s districts with different income levels? And (2) did these effects vary across product categories and over time, specifically between the short term (1–2 weeks) and the medium term (3–6 weeks)?
Materials and Methods
This study drew on three main data sources. The first comprised daily sales data for the Iranian calendar years 1400 (2021–2022) and 1401 (2022–2023) from the largest branch of a major retail chain in each of 16 of Tehran’s 22 districts. The retailer accounted for approximately 10% of retail purchases in Tehran and charged uniform prices across all its branches. The data covered seven product categories: groceries; meat; fruit and vegetables; dairy products and eggs; clothing; detergents, cosmetics, and personal hygiene products; and home appliances.
The second source was a 2% sample of the Iranian Refah Database, which provides individuals’ postal codes, economic deciles, and annual Shaparak card payments. The analysis used the median Shaparak payments per person in each district in year 1400 as a measure of district-level income. The third source was the Central Bank of Iran’s monthly Consumer Price Index (CPI) for 14 product categories over the period 1395–1401 (2016–2023).
The empirical analysis consisted of two stages. First, monthly inflation was forecast using an ARMA(1,1)–SARMA(1,1) model estimated over the pre-policy period. Product groups whose realized inflation in June 2022, immediately following the policy intervention, had fallen outside the 95% forecast interval were classified as experiencing an unexpected price increase. Second, a triple-difference-in-differences (DDD) framework was used to measure the effect of the policy package on consumption inequality. The analysis compared spring 1401 (2022) with the remainder of 1400 (2021) and 1401 (2022), exploiting variation across districts with different income levels and across months (or weeks) within the season.
The coefficient of interest measured the extent to which sales increased more, relative to the control period, in a district with 1% higher income. A positive value indicated an increase in consumption inequality. The main outcome variable was the logarithm of branch-level sales, and all specifications included branch and time fixed effects. The models were estimated separately for total sales and for each product category at monthly and weekly frequencies.
Results and Discussion
According to the results, the removal of the Tarjihi exchange rate was associated with heterogeneous price increases across product categories. Meat, dairy products and eggs, and groceries experienced unexpected inflation around the time of the policy intervention, whereas fruit and vegetables and the non-food categories did not exhibit comparable unexpected price increases.
Despite this price shock, the monthly analysis revealed no statistically significant change in overall consumption inequality. This finding also held when product groups were examined separately according to whether they had experienced an unexpected price increase. Thus, at the aggregate level, the policy package was not associated with a statistically detectable widening of consumption expenditure disparities across districts.
The disaggregated analysis, however, revealed substantial heterogeneity across product categories. The weekly estimates indicated a medium-term increase in consumption inequality for meat and dairy products and eggs following the policy intervention. In contrast, groceries exhibited no significant medium-term increase in inequality despite experiencing a price shock, possibly because they are essential goods with relatively price-inelastic demand. Fruit and vegetables, which did not experience a comparable unexpected price increase, likewise showed no policy-related increase in inequality; the upward trend in the monthly estimates predated the intervention. Clothing and home appliances also exhibited no significant medium-term effects, while detergents, cosmetics, and personal hygiene products showed only a modest increase in inequality.
The dynamic estimates further indicated that some product categories experienced a temporary decline in consumption inequality during the first one or two weeks following the intervention—particularly meat; dairy products and eggs; and detergents, cosmetics, and personal hygiene products. Given the monthly disbursement of cash transfers, this short-lived decline may reflect precautionary stockpiling of storable goods, particularly among lower-income districts. The absence of a similar pattern for fruit and vegetables, which are less amenable to storage, is consistent with this interpretation.

Table 1. Summary of Medium-Term Inequality Results





Product categories


Price increase


Medium-term inequality (3–6 weeks)




Groceries


✔


---




Meat


✔


↑




Fruit and vegetables


✖


---




Dairy products and eggs


✔


↑




Clothing


✖


---




Detergents/cosmetics/hygiene products^*


✖


---




Home appliances


✖


---





Note. ✔: realized inflation in Khordad 1401 above the 95% forecast interval; ↑: significant increase in inequality; ---: no policy-related increase. ^*Small increase in the weekly specification.
Conclusion
According to the findings, the policy package did not lead to a statistically significant change in overall consumption inequality. However, meat and dairy products and eggs—product categories that experienced substantial price increases following the reform—exhibited an increase in consumption inequality in the medium term. In contrast, groceries, fruit and vegetables, and most non-food categories showed no statistically significant policy-related increase in inequality.
These findings suggest that the distributional consequences of a multiple exchange rate reform cannot be assessed solely on the basis of aggregate consumption. The results are also consistent with cash transfers having mitigated the overall distributional impact of the reform, even as inequality increased in certain affected product categories.
Because the data was aggregated at the branch level, inequality within districts could not be measured. Future research can address this limitation by using household-level transaction data, such as Shaparak records, to examine within-district consumption inequality more precisely.