Over the last decade, it has become clear that Kenya’s National Hospital Insurance Fund (NHIF) is the vehicle through which the government hopes to eventually offer health insurance to all Kenyans. Since 2006, the fund has increased its membership of formal and informal sectors alike, and the share of national health resources that are under its direct management. The recent controversial attempt to expand outpatient coverage is also intended to position NHIF for a larger role in the health system.

However, if NHIF is going to anchor universal health coverage in Kenya, then it must be capable of managing its finances in an effective manner. Moreover, the Fund, like all state corporations, collects and spends public money and must report on its use of funds to permit effective oversight by Parliament and the public. State corporations in Kenya manage massive sums: approximately 610 billion KSh (US$7 billion) in FY 2011/2012, of which 139 billion KSh constituted government funds from the budget. Put into perspective, that is roughly 13 percent of budgeted expenditure for FY 2011/2012. Given their public role, state corporations like NHIF must be held accountable for the money they use, and this requires timely and transparent financial reporting.

Measuring Health Fund Transparency

In IBP Brief 14, we ask whether NHIF communicates financial and policy information to the public in a clear, consistent, and transparent form, as befits a major state corporation with ambitions to provide universal health coverage. Our answer is no.

In order to answer this question, we sought information from NHIF offices and the NHIF website between March and May of 2012. We looked for information that would answer five questions:

1. How much money does the Fund take in each year from its main activities, and how much does it spend on benefits?
2. Does NHIF receive subsidies from the government beyond the contributions employees make and if so, what is the amount received and how is it used?
3. How does the Fund manage its investments or expenditures other than member benefits?
4.  Is the Fund accountable for the money that it raises and spends?
5. What are official NHIF policies (Who contributes? How much? What benefits do they have access to?) and how are these changing over time in line with the Fund’s expanding role in the health sector?

The NHIF website contains no financial information about the Fund. And not one of these five types of information is available from the NHIF offices. After repeated requests over nearly two months, we were told we could have it only if we signed a strict confidentiality agreement! We did obtain information from other sources — difficult but not impossible to get — including the Kenya National Audit Office (KENAO). What we found reinforced the value of these five types of information for oversight.

For example, consider our third question on investments and other spending. Between 2004 and 2010, KENAO has raised concerns about the construction of NHIF’s multi-storey car park. The car park was originally contracted for about 900 million KSh in 2002. It was completed in 2008, but at a total cost of over 3.3 billion KSh. Then, an additional 626 million KSh was spent on the car park, resulting in a total expenditure of more than four times the original contract. As the auditor observes, “the escalation of costs…has not been justified, [and] the final completion certificate for the project had not been issued” two years after completion. This raises serious concerns about financial management and accountability at NHIF. NHIF also spends a very high share of revenues on personnel and administration: 45 percent in 2010.

Our brief argues that the lack of transparency at NHIF is representative of a broader problem of inadequate public reporting on the financial activities of Kenyan state corporations. For the last two years, the government has tabled an annex to the budget including estimates of revenue and expenditure for state corporations. However, this document is of limited utility, as it lacks critical information that would facilitate public oversight, and it is not available online.

What is to be done?

In light of the foregoing, we make a number of recommendations in this brief, including:

1. NHIF should immediately post on its website its audited financial statements for the past 10 years. This degree of transparency is not extraordinary for public corporations, even in Kenya. For example, the National Social Security Fund has posted audited financial statements from 2007 on its website.The annex to this brief contains other global good practice examples of health fund transparency. NHIF should also post online narrative information about its policies.
2. The Treasury should immediately post the annex on state corporations to its website, but should also revise it to include additional information. This would include much greater disaggregation of the income and expenditure of state corporations, a short narrative explaining each corporation’s business model, definitions of terms used, and an explanation of when blank spaces and dashes refer to zero income, and when they refer to unreported income.

In the era of structural adjustment, this debate played out in Africa and other developing regions. When these governments started running out of money, their ministries of finance would, often on advice from donors, determine ceilings within which each government department had to prepare its budget. In some cases these limits would also be apply to specific categories of expenditure such as salaries. Since the global financial crisis, the debate on budget caps has ironically shifted to Europe and the United States. But the question of how much is enough and how to curb unbridled growth in government spending is never far away. As soon as economies, and consequently government income, slows down, it returns to center stage.

 It ain’t what you do; it’s the way that you do it

Despite ideological differences, most people would agree that there needs to be some limit, however generous or stingy, on government spending. So the most interesting question about budget caps is not whether they should be used or not, but rather how they should be used. “The debate in New Zealand (reported on in the IMF’s PFM blog) has brought to the fore some innovate and flexible ways of working with budget caps.

The first important lesson to learn from New Zealand is that they don’t follow a one-size-fits-all approach, but rather use different kinds of caps for different categories of expenditure. Some caps make provision for adjustments for inflation and population growth. These kinds of caps can keep expenditure within broad limits, but also accommodate spending pressures that come from things like rising fuel prices or growth in learner numbers.

The Kiwis also distinguish between caps on capital and operational spending. The reason for this distinction is that operational spending is recurrent in nature. If more teachers are employed or a new social grant introduced, government is compelled to continue incurring these expenses over the medium to long-term. So if new operational spending commitments are made, government should also be sure of recurring tax and other revenue to finance this. By contrast infrastructure and other capital spending are once off in nature. Governments can therefore fund such spending from less renewable sources such as debt or revenue windfalls.

 Guidelines or laws?

The New Zealand debate has also focused on how rigid budget caps should be. Should they take the form of flexible guidelines or should they be written into the finance act? Interestingly the  PFM blog  refers to research that shows that in  countries where quantitative fiscal rules have been embedded in legislation, there have been many failures and few successes. Evidence from the State of Colorado suggests that the reductions in tax and spending that resulted from legally imposed budget caps had a negative impact on the state’s economy.  These reductions in expenditure resulted in worsening health and education outcomes between 1992 to 2005, when law was finally suspended. While legal limits may look like an attractive instrument for controlling spendthrift governments, their lack of flexibility tend to be harmful in the long run.

 Your experience?

What has your experience with budget caps been? Are any of the above ideas relevant to your country? Do you have any other ideas for how to work with budget caps in a flexible manner?

The difference between oil and non-oil states

Much like Wehner and de Renzio, Ross finds that among democracies, a country’s mineral wealth does not have much effect on the transparency of the government budget. Among autocracies though, greater oil wealth correlates with less fiscal transparency. So oil wealth makes the budgets of undemocratic countries less transparent. But contrary to what one might have expected, Ross finds that non-fuel mineral wealth makes government budgets more transparent.

The paper also demonstrates that oil has an effect on budget transparency that goes above and beyond its more general effect on democratic accountability. So there appears to be a version of the resource curse aimed at budget transparency specifically.

Why, you might ask?

Its not EITI

Based on the available data, Ross does not find convincing evidence that the different levels of budget transparency in countries dependent on oil and those dependent on other kinds of mineral wealth, can be explained by membership in the Extractive Industries Transparency Initiative (EITI). He argues that if EITI pressures were responsible for the exceptional transparency of some mineral-producing autocracies, one would expect that most highly transparent mineral autocracies would be EITI members. However in Ross’s analysis, of the five authoritarian mineral producers that are relatively transparent (South Africa, Russia, Botswana, Namibia and Zambia) – only one (Zambia) is a member of EITI. In fact, the mineral-producing autocracies that are EITI members (Zambia and Kazakhstan) have lower OBI scores than the mineral-producing autocracies than are not members, although the differences are not statistically significant.

It’s not foreign investment either

Ross also doesn’t find evidence that the transparency anomaly can be explained by dependency on foreign investment. Oil rich countries derive much more revenue from their extractive industries than countries rich in other minerals derive from theirs. The argument would go that non-oil countries would be more transparent because they are more dependent on foreign investment. But Ross doesn’t find any link between foreign investment inflows and budget transparency.

The clue lies in who extracts

Unsurprisingly Ross concludes that learning more about the different levels of budget transparency in oil and mineral dependent countries should be a priority for future research. He finds a promising clue to this puzzle in the fact that in most autocracies the state manages oil extraction itself through national oil companies, making it easier to cloak revenue. The extraction of other minerals is more often managed by the private sector. This still doesn’t explain why the regular budgets of oil states would also be less transparent. But the lower revenue derived by non-oil states may make them more dependent on domestic revenue such as taxes and service charges. The research of Mick Moore on taxation and democracy suggests that dependence on the consent of citizens to pay tax may compel governments to be more transparent with the management of public resources.

What does all this mean?

One of the important implications of Ross’s findings for the policy and advocacy community is that domestic politics, and therefore national level advocacy, matter a great deal. Like Wehner and de Renzio, Ross finds that the way in which extractives are managed at country level is the key variable for their impact on budget transparency. In fact, it may matter more than the need to attract foreign investment, and more than international advocacy campaigns. Of course this does not make international advocacy for transparency irrelevant. It just means that international advocacy efforts should be based on a better understanding of domestic contexts, and have a complementary focus on country level advocacy.

Many historical, structural and natural factors may play a role in the level of budget transparency in a given country. While relevant, research on such factors are less helpful to the policy and advocacy community, because they focus on things that can’t be changed. Knowing for example that francophone administrative heritage countries tend to be less transparent than anglophone heritage systems may hold academic interest, but doesn’t tell us how to make budgets more transparent.  For this reason also Wehner and De Renzio’s analysis is interesting: they look at determinants of budget transparency that can and do change.

Wehner and de Renzio’s first finding is that free and fair elections have a significant effect on budget transparency. Countries that have recently had free and fair elections tend to have more transparent budgets. Interestingly, their results also suggest that high levels of transparency do not rely on a slow process of democratic maturation that may take decades or even centuries. Any recent free and fair elections seem to be enough. The rapid improvements in the OBI scores of countries like Mongolia and Liberia  seem to support these findings.

Interestingly Wehner and de Renzio also find evidence that free and fair elections reduce the well-known adverse effect of natural resource dependency on budget transparency. Natural resource dependency is therefore not an inevitable death sentence to transparency.

Finally, Wehner and de Renzio obtain results that greater political competition is associated with higher levels of budget transparency. They find that the better the representation of opposition parties in the legislature, the higher the level of budget transparency.

While exploratory, this research suggests that proponents of budget transparency can promote greater transparency by supporting broader political reforms such as changes in the electoral system and other measures that enhance political competition. Along similar lines, more research that looks at the impact of media freedom on transparency would also be interesting.

Further, while it is helpful knowing that elections and political competition determinants are statistically associated with budget transparency, reformers and advocates also need to know how this happens. Just what does political competition and elections do to make budget transparency?

One of these papers provides convincing evidence that high levels of budget transparency can contribute to favorable credit ratings. But the links between budget transparency and credit ratings are sometimes not as straight forward as one might expect.

Financial markets have played an increasingly important role in the ability of both developed and developing countries to finance their budget deficits. Emerging markets in particular have succeeded in accessing credit to support their spending on infrastructure and to restructure existing debt on more favorable terms.

But before financing a country’s debt, lenders will assess the creditworthiness of the particular country so that it can set interest rates that will make the transaction profitable to them. In a recent research paper prepared for the Open Budget Initiative, Farhan Hameed finds that countries with higher levels of budget transparency are rewarded with better credit ratings. And the better the credit worthiness, the more favorable the terms of the debt.

Transparency shields against shocks

Hameed also finds limited evidence that more transparent countries are less likely to have their credit ratings downgraded because of sudden changes in market sentiment. Markets assess the merits of more transparent countries on the basis of known economic data rather than mere ‘market sentiment’.

Transparency can therefore help steady the boat of government finances in turbulent economic times, especially in smaller, outwardly oriented economies like South Africa that are more exposed to global market forces.

Medium transparency countries can benefit the most

But not all countries can access the rewards of budget transparency. Medium transparent countries like Tanzania, the Philippines and Mongolia stand to gain the most. Untransparent countries may not reap these benefits . If transparency brings to light adverse information, such countries may initially be punished by having their credit worthiness downgraded. Over the medium to long-term this effect should however diminish and these countries should also start reaping the rewards available to more transparent countries.

Highly transparent countries may also not reap the market benefits transparency improvements because they may have published enough budget information already for markets not to be impressed by further information.

Don’t jump to easy policy conclusions

Before we get too excited about the implications of this research, Hameed cautions against drawing easy policy conclusions . At this time, he writes, we cannot be completely sure about the nature of the link between transparency and financial markets. While it is likely that credit ratings are dependent on budget transparency, it is also possible that a more capable government may adopt transparent practices in order to benefit from better ratings. This would mean that improved credit ratings are caused by lenders’ perception of the capability of the government, rather than just of its budget transparency practices. In other cases, more transparent practices may be driven by other considerations such as political competition.  Future research should therefore investigate other determinants of transparency to better understand what brings about budget transparency and improved credit ratings.

More budget transparency research

Other themes dealt with in the OBI series of working papers does just this and provides greater insight into the other determinants of transparency such as natural resource dependency and administrative legacy. The OBI has also commissioned a series of case studies that go deeper into the drivers  and consequences of budget transparency in specific countries. The results of this research will be published in an edited volume by mid-2012. Watch this blog for further news!

PS Farhan Hameed was previously a researcher at the IMF and taught economics at Lahore University for Management Sciences. He is currently working with the Government of Canada.