At a first glance, developing countries’ reliance on consumption taxes, which are viewed as regressive, suggests a limited or negative impact on income redistribution via the tax system. However, there are two channels through which consumption taxes affect poor households differently than rich ones. The first is the “de facto” exemption of products purchased in the informal sector from taxation. The second is the “de jure” exemption of some necessity goods, in particular food items. The latter approach is often part of tax policies.
If households spend less (as a share of their budget) on a type of product as they become richer, exempting this type of product from taxation while taxing all other products will be progressive: the rich will spend a larger share of their budget on taxes than the poor. Exempting food from taxation would therefore enable governments to redistribute wealth to the poor if the poor spend a higher share of their budget on food than the rich. Similarly, the presence of a large informal (untaxed) sector in low-income countries would make consumption taxes progressive if the household budget share spent in the informal sector falls with income.
In a recent paper, we quantify the effect of both of these channels by assembling a new micro dataset of expenditure surveys from over 30 low-and middle-income countries that contains information on the types of stores at which purchases are made at the transaction-level. A major constraint in studying informality is that, by their nature, purchases are hard to observe and to link to consumers’ incomes. We take an innovative approach by using the type of stores in which households report purchasing items -- ranging from home production, to street vendors, to supermarkets -- to proxy for consumption from the informal sector.
This enables us to document the existence of a downward-sloping Informality Engel Curve: within each country, the informal budget share declines steeply with household income. The shape of the Informality Engel Curves implies that the de facto tax exemption of the informal sector makes consumption taxes progressive. Indeed, with a simple uniform rate on all goods, the richest quintile pays two times as much in taxes as the poorest quintile in the average country (Figure 1, blue line). Moreover, we find that exempting food items from taxation only slightly increases progressivity, once we take into account the informal sector (orange line).
Figure 1: Progressivity of Consumption Taxes (Average Across Countries)
We then build a simple model to understand what these consumption patterns imply for consumption tax policy. The existence of informal varieties of goods that cannot be taxed changes both the efficiency and equity (or redistributive) characteristics of consumption taxes. It increases the efficiency cost of taxing consumption because households can substitute to informal varieties when taxes increase. It makes consumption taxes progressive as long as Informality Engel Curves are downward-sloping. We calibrate the model for each country in our sample. A key finding is that in some of the poorest countries, taxing food at a lower rate relative to non-food cannot be justified on equity or efficiency grounds. This is because in these countries, poor households purchase most of their food from the informal sector and would benefit little from food being taxed at a lower rate.
A tax policy’s impact on inequality depends not only on its progressivity, but also on the level of tax rates and size of the tax base that determine the share of a household’s budget that must be devoted to paying taxes. We find that, after accounting for informal consumption, setting an (optimal) uniform rate on all formal consumption achieves a significant reduction in Ginis (first row, Figure 2 Panel A). This inequality reduction is much larger than existing estimates of inequality reduction from actual consumption tax policies in developing countries (first row, Panel B), which do not systematically account for the de facto exemption of the informal sector. In fact, setting a uniform rate on formal consumption achieves as much inequality reduction as the actual direct tax policies in developing countries (second row, Panel B). Moreover, after accounting for informal consumption, implementing a reduced rate on food products reduces inequality by a modest amount compared to the informal sector effect (third row, Panel A).
Figure 2: Reduction in Inequality from Different Tax Policies
Thus, consumption taxes are substantially more progressive in developing countries than many believe. Most of the redistributive gains from taxing consumption are due to the existence of large informal sectors and downward-sloping Informality Engel Curves; this result is robust to the indirect taxation of the informal sector in a value-added tax (VAT) system (for example through the use of formal inputs by informal firms). Policies designed to increase the progressivity of consumption taxes, such as taxing food products at a reduced rate, are, on the other hand, fairly ineffective redistributive instruments.
These results do not imply that efforts to reduce the informal sector should be abandoned. Rather, they caution that benefits from shrinking the informal sector should be weighed against potential equity costs. More generally, tax enforcement policies should take into account not only their impact on efficiency but also on inequality. Finally, in most countries, firms below a size threshold are exempt from taxation. This policy is often motivated by the large enforcement costs tax administrations incur when trying to tax these firms and the compliance costs to the firms themselves. The growing availability of digital technologies could lower these costs and make it possible to bring increasingly smaller firms into the tax net, removing the administrative rationale for exempting small firms from taxation. The results suggest that exempting small firms from taxation could still be justified on equity grounds