Key message 2 – Third-best policies, inefficient in developed countries, may prove efficient in developing countries

A ‘first-best’ approach to taxation assumes idealised conditions where governments face no barriers to information or enforcement. Under this approach, taxes are lump sum and based on an individual’s inherent ability or ‘type’.

As illustrated by the Danish case, even the most efficient systems face some informational barriers. Therefore developed countries have historically promoted ‘second-best’ approaches to improve the efficiency of tax collection in the face of information constraints. Commonly recommended policies include taxes on consumption, progressive income taxes, profit taxes, and value-added taxes (VAT) on goods and services.

Accepting that some informational barriers are inevitable, second-best policies aim to maximise social welfare with the instruments available, namely taxes on observable transactions (such as income and consumption) not ability. A central result in public economics is that secondbest approaches should promote production efficiency (Diamond-Mirrlees, 1971). Taxes on consumption, wages, and profits are encouraged while taxes on intermediate inputs, turnover, and trade are discouraged for distorting firms’ production decisions.

Optimal tax policies and instruments in developing countries may look very different to those in developed countries.

Both developed and developing countries use second-best taxes. However, the emphasis on production efficiency that characterises second-best policies ignores the constraints to enforcement and administration that face developing countries. In such cases, second-best approaches generate very low levels of revenue. Optimal tax policies and instruments in developing countries may look very different to those in developed countries.

In practice, tax structures of many developing countries incorporate policies that run counter to second-best approaches (Gordon and Li, 2009; Best et al., 2015). But in contexts with capacity constraints, such policies may prove better for revenue generation. For instance, where evasion on taxable profits is high, turnover taxes provide a broader and thus harder to evade tax base (Best et al., 2015). Production efficiency must be balanced against the need for revenue efficiency in developing countries.

Tax systems that account for differences in context and capacity in developing countries may produce very different policies. Where enforcement capacity and third-party verifiable financial transaction data is limited, policymakers should explore ‘third-best approaches’.

Let’s consider Minimum Tax Schemes (MTS), an example of a third-best policy commonly adopted by developing countries (Best et al., 2015). Here, firms are taxed either on profits or turnover, depending on which tax liability is larger.

Unlike profit taxes, turnover taxes cannot be evaded by over-reporting costs. In high evasion environments, turnover taxes may generate higher tax revenues than profit taxes.

One of the countries using MTS is Pakistan. A study conducted by IGC researchers in Pakistan demonstrates that turnover taxes could reduce evasion by up to 70% (Best et al., 2015). Gains to revenue efficiency from increased compliance significantly outweigh lost production efficiency. Rather than advocating that production-inefficient taxes be replaced by modern tax instruments, these results indicate that some production-inefficient taxes are better suited to countries with weaker enforcement capacities. Locally appropriate and optimal tax structures require a broader understanding of the trade-offs between production and revenue efficiencies.

Achieving higher tax revenues in developing countries is not simply a matter of increasing taxes – it requires careful consideration to identify optimal tax instruments. In developing countries, reforms to increase revenues must overcome barriers to collection and enforcement. Instruments must be flexible enough to adapt to changing public finance infrastructure and constraints.

Tax policy approaches: when third-best is best