The National Treasury recently released its draft Budget Policy Statement (BPS) 2017. The document came much earlier this year (it is normally tabled in Parliament by February 15 ) because of the need to finish the budget cycle well before the 2017 elections. However, its content is similar to previous years.

The budgetary process is meant to focus at the sector level (e.g., trade-offs between health, education, energy and so on) between roughly October and March each year, framed by the Budget Review and Outlook Paper (BROP) at the beginning of that period and the Budget Policy Statement (BPS) at the end. In many ways, the BROP is a draft of the BPS. Both documents offer proposals for the total size of the budget and its distribution across sectors.

The BROP is intended to initiate that discussion with a proposal that kicks off a set of sector hearings (to which the public is invited), while the BPS is meant to finalize the discussion with a final proposal submitted to Parliament.
If this is in fact the logic of the process, then we would expect the BROP to provide reasons for Treasury’s initial proposal. The BPS would illuminate us about the sector hearings that followed and how the initial proposal was modified to address concerns raised by sector stakeholders, including the public.

In general, the documents do not do this. The BPS in particular does not adequately acknowledge the process that began with the BROP and culminated with the tabling of the BPS. In recent years, the BPS has begun to mention public inputs, but there is still very little evidence that these inputs actually have any impact at all on the decisions taken in the document.

The bigger problem, however, is that the BPS does not actually speak to public concerns about the main question that it is supposed to resolve: how to distribute next year’s budget among a number of competing sectors, all of which have reasonable claims on government funding. The decision to allocate more to health or education or energy is what is at stake here, and the BPS offers little justification for government’s proposals at this level, and no insight into what the public actually thinks about this question.

Incompatibility of government’s and people’s preferences
With this in mind, my organization, IBP Kenya, teamed up with Infotrak Research and Consulting to carry out a nationally representative survey of Kenyans to ask them their preferences across the ten sectors of government. The survey was conducted in August 2016. What did we find?

At the most basic level, the survey finds significant differences between the sector preferences of the people and those proposed by the government. We asked members of the public to divide the total budget for sectors into the shares they would like to give to each. Here, we compare their responses to what the government has proposed for 2017/18 in the Budget Review and Outlook Paper 2016 issued last month.

We found that the public would give much less to energy and infrastructure than the government. While the government was allocating about 30% of the budget to this sector, the public only wished to allocate about 11%. A significant share of that difference would go to increasing the budgets for agriculture and health, by roughly 11 percentage points in both cases. In other words, the public would increase the agriculture budget by nearly 5 times what government proposed, and the health budget by 4 times the government’s proposal for 2017-18.

We also found that, in spite of strong support for funding education, the public would only give to education 80% as much as the government proposed, but would increase funding to the environment by 50% and would double the government’s proposed budget to social protection, which includes cash transfer programs.

How should we interpret this data? We considered three possibilities in our recent paper on the survey results. One is that we should take these numbers literally and in doing so, acknowledge that there is a massive gap between public preferences and government priorities. Assuming we take seriously the constitutional requirements for public participation in public finance, there would be a strong presumption that the government should revisit its priorities based on these results.

A second perspective is dismissive, suggesting that the public is not informed enough to form meaningful preferences about the sector allocation of the budget. In this case, we would not be as concerned about divergences between the public and government preferences, but the survey results would show that the government does very little to educate the public about the true costs of providing government services. This in turn would be inconsistent with the requirements of obtaining reasonable public inputs into the budget process and suggest that government must do much more to educate and persuade the public of its priorities.

A third perspective suggests that the public is not well informed enough to select specific figures for particular sectors, but that it does have stable preferences. In this case, we would interpret the survey results as “directional,” rather than literal, meaning that if the public says that agriculture should receive five times as much as government, what they mean is that the sector is more important than the government believes. Government should consider increasing the agriculture budget, but not necessarily by a multiple of five.

Which of these perspectives is most accurate? We do not believe the first perspective is sustainable, as there is a substantial amount of evidence, from our survey and elsewhere, that people are not well informed about the true costs of government. In fact, our survey asked people how much government currently spends on various sectors, and the results show that citizens are ignorant of the actual distribution.

This makes perspective two attractive, as it starts from the premise that people are uninformed. However, we must be careful in making the leap from the fact that people are uninformed to the claim that we should ignore their views. Information is only one input into people’s preferences, and preferences should be informed not only by technical knowledge but also by values. Given the many factors that shape preferences, we cannot assume that more information will or should change people’s views. While we strongly believe that more informed citizens do make better choices, we also believe that it is profoundly undemocratic to dismiss the views of the “uninformed” public. Our survey results suggest that even when respondents are given more information about the actual distribution of the budget they do not change their minds. Thus their preferences are more stable than one would expect if they were driven purely by information constraints.

This leaves us with perspective three, which allows that public views are partly uninformed, but holds that they nevertheless signal meaningful preferences about the direction of spending. This perspective, which we ultimately support, is bolstered by some of the discussions we observed in focus groups, where people were able to make reasonable judgements about trade-offs among sectors, even if they were not able to put precise numbers on them.
We conclude that the public believes that the government’s current budgets do focus too heavily on energy and infrastructure relative to agriculture, health, social protection and other areas. We do not, however, take a strong view on how much more focus the government should give to each of those areas relative to energy and infrastructure. We do not think a survey can reveal such precise information. But we do believe, based on our experience, that the public is capable of learning about and giving more precise answers over time.

The author is the head of the International Budget Partnership Kenya

The full analysis synthesized here is available at:
http://www.internationalbudget.org/wp-content/uploads/ibp-kenya-comparing-sector-priorities-public-government-2016.pdf