FISCAL POLICIES AND INSTITUTIONS FOR SHARED GROWTH IN KENYA

Transcription

1 Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized FISCAL POLICIES AND INSTITUTIONS FOR SHARED GROWTH IN KENYA LESSONS FROM THE GLOBAL CRISIS Public Disclosure Authorized APRIL 2010

2 Note: Fiscal years, currency and equivalent units Fiscal year: 1 July-30 June Currency = Kenyan Shillings (Ksh) US$ 1.00 = Ksh Abbrevia ons and Acronyms EBCs OPM TJRC ASAL BOP BoT BPO BSD BSP CBK CCT CEM CPI DFID DOTS DPT EBSC ECV EMOP ERD EWS FY GDP GFS GJLOS GoK HIV/AIDS HP HSNP ICT IDA IDA IMF JICA KFSG KIHBS KJAS Km KNBS KPIA Electronic Benefit Cards Office of the Prime Minister Truth, Jus ce and Reconcilia on Commission Arid and Semi- Arid Lands Budget Outlook Paper Build, Own, Operate and Transfer Business Process Outsourcing Budget Supply Department Budget Strategy Paper Central Bank of Kenya Condi onal Cash Transfer Country Economic Memorandum Consumer Price Index Department for Interna onal Development Direc on of Trade Services Diphtheria, Pertussis, Tetanus Economic and Budget Steering Commi ee European Community Emergency Relief Opera ons External Resources Department Early Warning System Fiscal Year Gross Domes c Product Government Finance Sta s cs Governance, Jus ce, Law and Order Sector Government of Kenya Human Immuno- Deficiency Virus/AIDS Hodrick-Presco Hunger Safety Net Program Informa on and Communica on Technology Interna onal Development Agency Interna onal Development Agency Interna onal Monetary Fund Japan Interna onal Coopera on Agency Kenya Food Security Group Kenya Integrated Household Budget Survey Kenya Joint Assistance Strategy Kilometer Kenya Na onal Bureau of Sta s cs Kenya Poverty and Inequality Assessment Ksh LIC MDER MDGs MNKOAL MOA MOE MOED MOF MOGC&SD MOR MOWI MPER MTEF MTP MVC NAAIAP NARC NCPB OVC PAYE PEFA PIM PIU PPPs PSNP SAS SMEs SSA SWGs UNICEF VAR VAT WB WDI WFP Kenya Shillings Low Income Countries Minimum Dietary Energy Requirements Millennium Development Goals Ministry of Northern Kenya and Other Arid Lands Ministry of Agriculture Ministry of Energy Ministry of Educa on Ministry of Finance Ministry of Gender and Children and Social Development Ministry of Roads Ministry of Water and Irriga on Ministerial Public Expenditure Review Medium Term Expenditure Framework Medium-Term Plan Most Vulnerable Children Na onal Accelerated Agricultural Inputs Access Program Na onal Rainbow Coali on Na onal Cereals and Produce Board Orphaned and Vulnerable Children Pay As You Earn Public Expenditure and Financial Assessment Public Investment Management Project Implementa on Unit Public- Private Partnerships Produc ve Safety Net Program Seasonal Adjustments Small and Medium Enterprises Sub-Saharan Africa Sector Working Groups United Na ons Children s Fund Vector Autoregressive Value Added Tax World Bank World Development Indicators World Food Program

3 Table of Contents EXECUTIVE SUMMARY i A. Fiscal Policies and Ins tu ons for Shared Growth i B. Characterizing Kenya s Public Investment program ii C. Automa c Stabilizers, Counter-Cyclical Transfers to Protect the Vulnerable iv D. Conclusions vi 1. FISCAL POLICIES AND INSTITUTIONS FOR SHARED GROWTH 1 2. A GLOBALLY COMPETITIVE AND PROSPEROUS KENYA: THE ROLE OF PUBLIC INVESTMENT PRO-POOR SPENDING IN KENYA: A REVIEW OF THE TARGETED SUBSIDY PROGRAMS 36 References 49 Appendix 50 Appendix Group A: Fiscal Policies and Ins tu ons for Shared Growth 51 Appendix Group B: A Globally Compe ve and Prosperous Kenya-The Role of Public Investment 65 Appendix Group C: Pro-poor Spending in Kenya: A Review of Exis ng Targeted Subsidy Programs 69 List of Boxes Box 1: Public investment: Does it crowd in or crowd out private investment 13 Box 2: Public investment management: A note on good prac ce 24 Box 3: Vision 2030 and the New Coali on Government 29 Box 4: Vision 2030, the medium term plan and spending needs 30 Box 5: Disbursement problems: The case of World Bank 33 List of Figures Figure 1: Economic performance, i Figure 2: Correla on between budget surplus and cyclical component of GDP ii Figure 3: Cyclicality of Kenya s development budget iii Figure 4: The external sector is the main source of vola lity in Kenya s GDP iii Figure 5: Composi on of development budget iv Figure 6: Kenya s infrastructure lags behind other low income countries in Africa iv Figure 7: On average, only two thirds of the development budget is u lized each year, iv but the ra o has been increasing Figure 8: Targeted spending accounts for about 1 per cent of GDP, mainly donor funded v Figure 9: Provincial distribu on of service gains, /06 15 Figure 10: Reduced distances to various infrastructure, Figure 11: Rise in development spending, 1999/ /07 18 Figure 12: Acquisi on of non-financial assets, Kenya and comparators, Figure 13: Share of development spending by sector, per cent of total 2004/ /09 20 Figure 14: Ministerial Spending 2000 Prices: Spending shi to physical infrastructure 21

4 Figure 15: Total donor commitments 2008/09 22 Figure 16: Government departments with donor-financed projects, 2008/09 budget 23 Figure 17: Budget devia on by ministry 28 Figure 18: Top 5 under-spenders 28 Figure 19: Sectoral budgets 2008/09 and 2009/10 31 Figure 20: Regional poverty levels (KIHBS 2005/06) 37 Figure 21: Prevalence of cri cal food poverty (KIHBS 2006/06) 38 Figure 22: Targeted programs popula on and shares of total expenditures 2008/09 (percent) 44 Figure 23: Spending on different categories of vulnerable groups 46 Figure 24: Regional poverty (numbers 000) and expenditure per capita 47 Figure 25: Poverty incidence and total targeted spending by province 47 List of Tables Table 1: Factors explaining decline in indebtedness, 1996/ /07 (percent of GDP, annual average) 2 Table 2: Central government fiscal outcomes as percent of GDP, 1999/ /07 3 Table 3: Correla ons among various measures of the cyclical component of GDP, Q3 4 Table 4: Summary budget figures, Table 5: Components of public sector revenue and expenditure, Table 6: Cyclical proper es of public sector revenue and expenditure, 1996 Q Q3 6 Table 7: Es mates of revenue elas ci es 7 Table 8: Correla on of the budget surplus measures with the cyclical component of GDP 8 Table 9: Variance decomposi on of output (percent of variance of the forecast error) 11 Table 10: Public investment outcome indicators, Kenya in comparison, Table 11: Indicators of human capital, Kenya and SSA and LIC averages 15 Table 12: Economic composi on of development budget, 2004/ /09 as percent of total 17 Table 13: Fiscal space for increased development spending 2002/ /09 18 Table 14: Kenya and comparators, development spending as percent of GDP 20 Table 15: Share of development budget by sector, 2004/05 to 2008/09 (Ksh millions) 20 Table 16: Sectoral composi on of development expenditures (Ksh millions) 22 Table 17: Donor-financed development expenditures, 2002/ /09, Ksh billions 22 Table 18: Development budget execu on rates, 1999/ /07 26 Table 19: Budgeted and disbursement varia on, 2006/07 US$ millions 29 Table 20: Targeted schemes in Kenya: Popula on coverage (numbers 000) 29 Table 21: Selected targeted programs in Kenya 40 Table 22: Programs with poten al for scale up 42 Table 23: Expenditure levels by program, 2008/09 (Ksh million) 43 Table 24: Fiscal opera ons (Ksh billions) 44 Table 25: Regional spending social safety nets 45 Table 26: Average government transfer in Kenya by province and poverty status 47

5 Acknowledgements This report was prepared by a core team led by Jane Kiringai (AFTP2), together with Cevdet Denizer (CF- PIR) and Tracey Lane (AFTP2). Important substan ve comments were received from Jeni Klugman (AFTP2), Sanjay Dhar (AFTPM), Harold Alderman (PA9SS), Luis Serven (DECRG) and Sudharshan Canagarajah (EC- SPN). Special thanks go to the consultants Fredrick Wamalwa, Owen Nyangoro, Lydiah Ndirangu, Njeru Kirira and Yuliya Meshcheryakova for excellent research and data analysis. A larger group within the World Bank contributed valuable inputs to this report, for which the core team expresses their apprecia on: Wolfgang Fengler (AFTP2), Louise Fox (AFTP1) and Mike Mills (AFTHE) who offered important insights and comments for the ini al dra s, and Carolyn Wangusi and Anne Kha mba (AFTP2) who provided editorial and administra ve assistance. The team also expresses their apprecia on to the Targeted Food Subsidy Scheme Task Force for the invaluable data used for the analysis. We are also grateful for the contribu ons provided by the Office of the Prime Minister, Ministry of Gender and Children Affairs, Ministry of Educa on, Ministry of Agriculture, and World Food Program. Overall guidance was provided by Kathie Krumm (Sector Manager AFTP2).

6 EXECUTIVE SUMMARY This report brings together three budget notes that assess Kenya s fiscal capacity to respond to global crisis and deliver shared growth without compromising macroeconomic stability. The key messages from this analysis point to areas where Kenya s fiscal policy requires strengthening. This Execu ve Summary summarizes messages from three budget notes. The first note, Fiscal Policies and Ins tu ons for Shared Growth, is an assessment of Kenya s fiscal stance and suggests an appropriate fiscal anchor for Kenya. The second note, A Globally Compe ve and Prosperous Kenya: The Role of Public Investment, reviews the status of public investment in Kenya and suggests the reforms required to improve public investment planning and implementa on. The subject of the third note is Pro-poor Spending in Kenya: A Review of the Targeted Subsidy Programs. This study suggests that the targeted cash transfers, unemployment benefits, and workfare programs provide automa c stabilizers for fiscal policy during crisis. This Execu ve Summary ends with some concluding recommenda ons on the opportuni es for strengthening fiscal policy. A. Fiscal Policies and Ins tu ons for Shared Growth Introduc on Kenya s strong economic performance between 2002 and 2007 has been partly a ributed to macroeconomic stability and strong fiscal consolida- on. A er two decades of sluggish performance, economic growth resumed in 2002 and steadily increased from 0.5 per cent to 7 per cent in 2007 (see Figure 1 below). During this period, the government re red debt and started crea ng fiscal space to fund essen al infrastructure. The ra o of debt to Gross Domes c Product (GDP) declined from 60 per cent in 2000 to 40 per cent in Fiscal space was achieved through a strong revenue effort and stringent fiscal management. The budget deficit averaged about 2 per cent during this period. As a result of the strong growth performance, income per capita increased and poverty declined from 56 per cent in 2000 to 47 per cent in 2005/06. Figure 1: Economic performance, GDP growth rate Public Debt% of GDP Source: Budget Speech 2009 i

7 Execu ve Summary This remarkable growth was disrupted by three shocks, star ng with the poli cal crisis experienced in the first quarter of 2008, the global financial crisis, and the prolonged drought running through to the third quarter of The combined effect of these shocks is manifested through sluggish economic performance, with GDP growth at 1.7 per cent in 2008, and about per cent in Income per capita contracted by 1.1 per cent in 2008, and it is es mated that close to 6 million Kenyans required food assistance as a result of the prolonged drought and high food prices. Although the government had created fiscal space to finance essen al infrastructure and social spending, the addi onal financing to counter the effects of the drought and global crisis will impart significant pressure on the available resources. The global financial crisis and the nega ve domes- c shocks present challenges and opportuni es as well. For Kenya, the introduc on of a targeted cash transfer program and a youth workfare program are good examples of the opportuni es. If well designed, these social programs will provide in-built automa c stabilizers for Kenya s fiscal system, which could be scaled up during crisis. In addi on to the social programs, a fiscal s mulus package equivalent to 1 per cent of GDP is also being implemented in response to the crisis. Characterizing Kenya s Fiscal Policy Surprisingly, unlike most developing countries, Kenya s fiscal policy is acyclical and does not lean with the wind. This finding is significant because most developing countries adopt a pro-cyclical fiscal policy, contribu ng to macroeconomic vola lity. Figure 2 below shows the correla on between budget surplus and the cyclical components of GDP between two periods, and The primary balance is posi vely correlated with GDP (0.08), thus it is higher during booms and lower during recessions, sugges ng counter-cyclicity. However, for the period , the coefficient dropped to 0.02, sugges ng that fiscal policy had become acyclical. Since the current government took office in 2003, fiscal policy has been me consistent. There have been no major differences between policy announcement percentents and outcomes, and the overall macroeconomic framework has remained broadly consistent with targets. This posi ve performance can be a ributed to the strong fiscal adjustment between 2002 and 2007, with remarkable reduc on in debt as a share of GDP and a strong revenue effort. Public borrowing was limited to 1.8 per cent of GDP. These efforts have paid off, and Kenya can now issue debt at single digit interest rates. Figure 2: Correla on between budget surplus and cyclical component of GDP Overall Balance Primary Balance Adjusted Overall Balance Adjusted Primary Balance Source: World Bank staff es mates ii

8 Execu ve Summary Public investment, other recurrent spending, and corporate taxes exhibit counter-cyclical proper- es, but development spending is highly vola le. Figure 3 below shows the correla on between fiscal variables and the cyclical component of GDP for two periods, and The low correla on coefficients for most variables suggest that fiscal variables are neutral to swings in the business cycle. During the first period, , pension payments exhibited counter-cyclical proper es but, since 2003, when the Na onal Rainbow Coali on (NARC) government took office, a new set of variables, development and other recurrent spending have become more counter-cyclical. Development spending remains extremely vola le at 50 per cent variability compared to 8 per cent for other recurrent spending. Figure 3: Cyclicality of Kenya s development budget Source: World Bank staff es mates Key: Dev-Development, Rec-recurrent, Inc. Tax Corp-Corporate Income Tax, Ims-Imports, T-Total, VAT_L-Vat Local, Ex-Excise, Int-Interest, It is Kenya s monetary policy, and not its fiscal policy, that has a significant impact on GDP s vola lity. The main sources of vola lity are in oil price shocks, and other external shocks transmi ed through the exchange rate. Figure 4 below shows that fiscal GDP accounts for 13 per cent of vola lity in output, while the federal funds rate accounts for 70 per cent of vola lity. In this regard, the federal funds rate is used as a proxy of the global market. Overall, fiscal shocks have limited impact on GDP vola lity; a 10 per cent change in cyclically adjusted primary balance leads to a 1 per cent change in GDP. Figure 4: The external sector is the main source of vola lity in Kenya s GDP Source: World Bank staff es mates In summary, the analysis suggests that Kenya s fiscal policy is not the cause of vola lity in GDP and, therefore, does not lower growth. As men oned earlier, vola lity in Kenya s output emanates from monetary policy, and notably from the exchange rate. Since the NARC government took office, development spending has become appropriately more counter-cyclical, but remains vola le. This is certainly a posi ve performance rela ve to developing countries whose fiscal policy responses have been pro-cyclical. However, there is scope for further reforms that would make fiscal policy counter-cyclical to complement the posi ve macroeconomic developments. B. Characterizing Kenya s Public Investment program Although total government expenditure accounts for about 27 per cent of GDP, development spending has averaged 4 per cent of GDP, and only 2 per cent of GDP is used for the core assets. On a posi- ve note, development spending has been rising in recent years; in 2008/09 it increased to 6 per cent of GDP. However, a closer analysis of this expenditure (see Figure 5) shows that, on average, 48 per cent of development spending is used for the acquisi on of assets, with the rest used for recurrent type expenditures. This implies that acquisi on of non-financial public assets has been about 2 per cent of GDP. However, it is notable that a significant share of development spending is off-budget, especially through state-owned enterprises. The expenditure analysis here is limited to Central government spending and excludes expenditure by state-owned enterprises, which will be covered in the more detailed 2010 Public Expenditure Review. iii

9 Execu ve Summary Figure 5: Composi on of development budget Source: Ministry of Finance, Notes constant 2004 prices The challenges in Kenya s public investment program are reflected in the state of infrastructure. Kenya s infrastructure development lags behind other low income countries in sub-sahara Africa (SSA) (see Figure 6 below). The most significant difference is in the paved roads density, where Kenya has a density of 16km compared to a SSA average of 31km, and an average of 134km for other Low Income Countries (LIC). Electricity coverage also lags, with a popula on coverage of 18 per cent in Kenya compared to 72 per cent for middle income countries. Even the country s 2011 target coverage of 33 per cent will be below the current average for LICs. Poor infrastructure, notably in transport and electricity, constrains economic growth and has been iden fied as a major performance challenge under the Doing Business Indicators. The good news is that the addi onal fiscal space created through debt re rement has been u lized to increase infrastructure spending. This increase in development spending has been brought about by the fiscal space created from an increase in revenue from 21 per cent of GDP in 2003/04 to 24.4 per cent of GDP in 2008/09, and restraint on the growth of recurrent expenditures (maintained at an average of 20percent of GDP over the same period). However, the increase in infrastructure spending also reflects a trade-off largely at the expense of the budget shares to the health, educa on and governance sectors. The execu on rate of the development budget has been improving from around 40 per cent in 2001/ 02 to over 70 per cent by 2006/07, but it s ll lags behind total budget execu on. On average, one third of development budget is not spent each year (see Figure 7 below). Donor financing contributes to poor execu on of development budget through delays in disbursements, but the problem can also be traced to over-budge ng and weaknesses in repor ng. As the government mobilizes domes c resources, the share of donor-financed projects has been declining and, between 2002 and 2008, the share of donor-financing declined from 73 to 41 per cent. Figure 7: On average, only two thirds of the development budget is u lized each year, but the ra o has been increasing Figure 6: Kenya s infrastructure lags behind other low income countries in Africa Source: Yepes et al. (2008), WDI (2008 ) 0 Source: Ministry of Finance (as reported in the Concept Note), 2007/08 World Bank staff es mates Weak budget execu on increases the lead me to project comple on, leading to escala ng costs. In addi on to low execu on rates, weaknesses persist in the project planning process. For instance, there are no procedures or evalua on criteria for appraisal and screening to ensure that high-return projects are priori zed. As a result, some of the projects are poorly designed, which delays implementa on. Furthermore, future recurrent cost implica ons of such projects are not taken into account, leading to inadequate provision for opera ons and maintenance. iv

10 Execu ve Summary Improvement in development budget execu on is key to improving the ability of fiscal policy to respond to exogenous shocks. By crea ng fiscal space for infrastructure financing, Kenya has made remarkable progress in reversing the trend of low levels of public investment. Kenya is implemen ng a fiscal s mulus package equivalent to 1 per cent of GDP in response to the global financial crisis. The s mulus package has a strong focus on rural infrastructure and social spending. However, weak implementa on capacity and low absorp on of development funds could neutralize the poten al impact of the counter-cyclical s mulus. Accelera ng investment budget execu on and implementa on, and in par cular on projects with aid financing commitments, can complement the s mulus package. This approach would also have beneficial impact on the balance of payments. For instance, increasing development budget execu on from 50 to 80 per cent in fiscal year (FY) 2009/10 would increase total spending by an addi onal US$ 400 million, without addi onal debt commitments. C. Automa c Stabilizers, Counter-Cyclical Transfers to Protect the Vulnerable Kenya has a high number of poor and vulnerable individuals and households, but social safety nets to cushion the poor are limited and fragmented. Es mates based on the 2005/06 integrated household budget survey indicated that about 47 per cent of Kenyans live below the poverty line. Further, es mates indicate that about 15 per cent of the urban popula on and 22 per cent of the rural popula on live below the cri cal food poverty line. These es mates were based on a good year when the Kenyan economy was growing at about 6 per cent, and incomes per capita expanding at 3.0 per cent. However, a er the three shocks hit the country, incomes per capita contracted by 1.1 per cent, meaning that the number of poor and vulnerable Kenyans will increase un l economic growth exceeds popula on growth on a sustained basis. The impact of the crisis poses significant challenges towards the achievement of the Millennium Development Goals (MDGs) and, if not addressed, can reverse the gains made so far. Global experiences show that well designed, targeted countercyclical safety nets prevent long term consequences of poverty, such as hunger and malnutri on, which are especially damaging to children. The global es- mates indicate that spending on social safety nets varies substan ally by country, but averages about 1.9 per cent of GDP for both cyclical and countercyclical transfers. The current review shows that Kenya is currently spending about 1.0 per cent of GDP, including external funding (see Figure 8 below), but it is mainly ad hoc and not counter-cyclical by design. Figure 8: Targeted spending accounts for about 1 per cent of GDP, mainly donor funded Source: World Bank staff es mates Kenya has a significant number of programs funded and managed by different agencies, and mainly donor supported, but with limited coordina on. Consequently, there is no program that the government could easily and quickly scale up to cushion the vulnerable against crisis. Kenya has more than 14 targeted subsidy programs, but the geographic coverage is limited and skewed in favor of droughtprone rural geographic areas, with limited a en on to other vulnerable groups, par cularly the urban poor. The Government introduced a generalized maize subsidy scheme in November This approach was expensive and had to be suspended. The Government is now in the process of designing a new cash transfer program, and made a provision of Ksh 7.6 billion (US$ 100 million) in the fiscal year 2009/ 10 budget for social safety nets and a youth workfare program. Global experience shows that the design of a comprehensive and effec ve targe ng system takes at least nine months, some mes even longer, to put in place. Furthermore, even in countries that have counter-cyclical safety nets in place, scaling up in mes of crisis can pose significant challenges, especially when fiscal space is limited. v

11 Execu ve Summary With the high number of poor people in Kenya, it is unlikely that a subsidy scheme can comprehensively cover all the poor and vulnerable within the limited fiscal space, and without compromising the hard earned fiscal space. In addi on, the coverage and design of the scheme may require pu ng into considera on spa al differences in inequality and poverty, which has become a source of tension and fragility. Overall, the crisis has presented opportuni- es and challenges and, in this regard, has opened up policy dialogue, providing a good opportunity for a more coordinated approach between government and donors in the design of social safety nets. Preparing the groundwork for such programs now could help ensure that they can be scaled up later to respond to the next crisis or nega ve shocks. D. Conclusions Kenya has made good progress in macroeconomic management, which is benchmarked on a debt to GDP ra o as the fiscal anchor. This approach has paid off; Kenya is able to finance a counter-cyclical fiscal s mulus from the domes c market at single digit interest rate without compromising macroeconomic stability. Going forward and as the economy recovers, adherence to a clear exit strategy for the fiscal s mulus program would ensure that Kenya quickly reverts to a credible and sustainable fiscal stance, and the economy is once again on a sustainable growth path. The development budget s planning and its implementa on is the Achilles heel in Kenya s fiscal management. In the short term, increasing the rate of budget execu on remains the priority. However, for the medium to long term, a government system for project evalua on and selec on criteria will be necessary to help priori ze and sufficiently fund those projects with the highest rates of return and, thus, ensure that they also play a cataly c role in Kenya s growth. The global crisis has revealed that in addi on to addressing the needs of the vulnerable, social safety nets can be scaled up and will enhance the counter-cyclical role of the budget. However, to ensure fiscal sustainability, a degree of modera on is warranted for the schemes and, in par cular, stronger alignments with complementary long term development strategies for poverty reduc on is crucial. vi

12 1. FISCAL POLICIES AND INSTITUTIONS FOR SHARED GROWTH Introduc on The post-elec on violence in early 2008 has shown that when poverty and inequality provide a platform for poli cal contest, the risks emana ng from ethnic and social tensions are indeed real. As a result of the violence and security-related issues, all sectors suffered declines in value added, and GDP growth was li le over 2 per cent in The crisis added to government expenditure, and early es mates indicate that the budget deficit for the year 2008/09 was around 5 per cent of GDP, almost double the amount of previous year. The turmoil in interna onal markets, and the rising energy and commodity prices brought new challenges to policy management. Infla on, mostly due to external factors, has increased and the exchange rate has come under some pressure, though recent data suggests that there is a degree of stability in foreign exchange markets in Kenya now. While the impact of the ongoing interna onal credit crunch may not have hit Kenya in a pronounced and direct way, indirect impacts have manifested themselves in the form of reduced capital flows, slowdown in exports, migrant remi ances, reduced access to interna onal credit markets for Kenyan firms and government, and lower tourism revenue. Collec vely, these show that the favorable external environment Kenya enjoyed un l late 2007 has been affected, and is no longer a posi ve factor suppor ng economic growth. The government is well aware that the current global financial crisis puts a premium on sound fiscal policy management and, in this regard, there are significant cross-cu ng fiscal issues from the budget, the public investment plans, and the poverty points of view. Sound fiscal policy has two major complica ons for government. First, is that although the government has adopted debt to GDP ra o as a fiscal anchor, the indicator does not factor in the cyclicality of GDP. This suggests that a cyclically adjusted deficit would be a useful tool for government and, in fact, would help Kenya manage its na onal debt in a more appropriate way. The second complica- on is the potency of fiscal expansion to boost economic growth, and the ra onale of a fiscal s mulus. While this issue has been under discussion in Kenya since early 2009, there has been no formal analysis of the ma er. One key objec ve of this note is to provide such an analysis to be er inform the ongoing debate. Given the government s interest in issuing a Eurobond (though it did not take place in 2009) to finance large infrastructure needs, the Interna onal Monetary Fund (IMF) has raised the issue of establishing a ceiling on both domes c and external borrowing as the new fiscal anchor. In agreement, the government has since adopted a total debt to GDP as the fiscal anchor, with a medium term target of 40 percent of GDP. This budget note is the first in a 1

13 Fiscal Policies And Ins tu ons For Shared Growth series of three notes and it analyzes Kenya s fiscal stance, and suggests an appropriate fiscal anchor in the context of a more challenging post-crisis environment. The note starts with an overview of the fiscal situa on in Kenya and then uses two econometric approaches to gauge Kenya s fiscal stance. In conclusion, the note observes that unlike most developing countries, Kenya s fiscal policy is acyclical and does not lean with the wind, which is a posi ve outcome as a result of fiscal retrenchment during the last decade. Key Fiscal Policy and Growth Issues Overview of Fiscal Situa on in Kenya Kenya s debt to GDP ra o fell by over 30 points during the period 1995/96 to 2006/07. It is es mated that the current level of total debt stock is around 40 per cent of Kenya s GDP, with the share of foreign exchange denominated debt being at 60 per cent. This is a remarkable fiscal adjustment, which had the effect of reducing country risk, which in turn led to declines in real interest rates. This situa on, combined with sustained revenue effort, enabled Kenya to lay the founda on for the solvency of the public sector. As a result, the country has been able to issue debt at single digit interest rates, a situa on not enjoyed by larger emerging market economies. This policy credibility, which paid par cular a en- on to the level of domes c debt a net domes c financing posi on has been the main fiscal indicator for the government. It has allowed the private sector to expand its horizons, and encouraged it to invest in produc ve ac vi es. The end result has been the steady rate of growth of Kenya s economy during the period which, however, could not be sustained in 2008 due to the combined effects of violence that followed the December 2007 elec ons, and the ongoing global financial crisis. A closer look at factors that explain the decline in Kenya s debt to GDP ra o is provided in Table 1 above. By and large, all the five factors worked together in most of the three periods, and total debt stock rela ve to GDP declined. The most significant factors have been real GDP growth and real apprecia on of the exchange rate. What then can be expected within the next few years? Star ng from the current year, 2009, and making some assump- ons, it is possible to deduce some forward looking es mates. As noted already, economic growth has slowed down and, given the current situa on in interna onal markets, it is reasonable to assume that economic growth rates will not be as high as in the period. One of the perceived risks at the beginning of the global financial crisis was the poten al deprecia on of the real exchange rate. However, Kenya has weathered the global financial crisis rather well and, as a result of adop ng a new methodology for measuring infla on, real interest rates have been posi ve since 2009 and the exchange rate has stabilized against the dollar. Table 1 shows the evolu on of overall fiscal outcomes. Primary balance was in surplus during most of the past seven years, though its size has been declining. Table 1: Factors explaining decline in indebtedness, 1996/ /07 (percent of GDP, annual average) Change in public sector debt Contribu on from: 1. Primary deficit (- surplus) Real GDP growth Real interest rate Real exchange rate (- apprecia on) Other factors Source: CEM: Accelera ng and Sustaining Inclusive Growth, July / / / / / /07 2

14 Fiscal Policies And Ins tu ons For Shared Growth Table 2: Central government fiscal outcomes as percent of GDP, 1999/ /07 Source: Interna onal Monetary Fund (IMF) 1999/ 2000/ 2001/ 2002/ 2003/ 2004/ 2005/ 2006/ Revenues Expenditures Recurrent Development Overall balance Primary balance Financing Domes c External Stock of domes c debt, net As shown above, expenditures have been increasing in line with GDP growth. In fact, expenditures have been growing at a higher rate than GDP growth, and the share reached almost 25 per cent in from 19.5 per cent in Fiscal Policy and Macroeconomic Stability Interna onal experience and empirical research show that fiscal policy tends to be expansionary during booms and contrac onary during recessions, especially in developing countries. In other words, it tends to lean with the wind. Therefore, in developing countries, fiscal policy is likely to exacerbate business cycle fluctua ons, rather than smooth them. A pro-cyclical fiscal policy, generally speaking, is undesirable and can be eliminated in two ways: first, by crea ng strong automa c stabilizers (for example given a fixed tax rate, tax revenues usually fall in recessions and increase in booms, while unemployment benefits rise in recessions), and second, by adjus ng the discre onary components of the budget. The cyclicality of fiscal policy and budget balance, in par cular, is only an issue if output responds strongly to changes in fiscal policy. Ilzetzki and Vegh (2008) show that it is indeed o en the case for developing economies. Focusing on the cyclicality of Kenya s fiscal policy, this note considers two issues: first, whether Kenya s fiscal policy indeed leans with the wind and, second, the effect Kenya s fiscal policy has on the country s GDP. The study uses quarterly observa- ons from 1996 Q1 to 2008 Q3, the real GDP, and Consumer Price Index (CPI) series compiled by the Kenya Na onal Bureau of Sta s cs (KNBS) and the Central Bank of Kenya (CBK), and the fiscal accounts data from Quarterly Economic and Budgetary Reviews compiled by Kenya s Ministry of Finance. Real exchange rate comes from the CBK. We use the CPI series rather than the GDP deflator to convert fiscal accounts into constant prices. This decision is driven by the fact that the KNBS started compu ng the GDP deflator only in 2000 and, therefore, the data is only available at annual frequency before The methodology and structure of the paper is as follows. First, cyclicality of Kenya s GDP and fiscal accounts are analyzed. We then compute and analyze Kenya s cyclically adjusted budget surplus. Further, we characterize Kenya s fiscal policy and observe that, in fact, it does not appear to be leaning with the wind, as one might have expected it to. In the last sec on, we construct and es mate a small vector autoregressive (VAR) model of Kenya s economy, which allows us to evaluate the impact of Kenya s fiscal policy on GDP. The analysis suggests that a 1 per cent increase in Kenya s cyclically adjusted primary balance causes GDP to decline by less than 0.1 per cent. In other words, while fiscal ghtening causes GDP to contract, this effect is fair- 3

15 Fiscal Policies And Ins tu ons For Shared Growth ly weak, and factors other than fiscal policy might be more important determinants of GDP fluctua- ons. This finding also implies that predic ng the potency of the impact of a fiscal s mulus package will be difficult, although the component of fiscal policy studied in this paper suggests a small impact. Results suggest that the US Federal Funds rate and Kenya s monetary policy might be among the factors determining GDP fluctua ons. However, this result must be interpreted with cau on; more research is needed in the case of Kenya before a defini ve conclusion can be arrived at. VARs of this nature deal with fiscal innova ons (purely unpredictable fiscal policy) and ignore two other important fiscal policy components: (i) systemic discre onary policy; rou- ne responses to changing economic condi ons; and (ii) automa c policy dictated by laws and rules such as the tax code. Business Cycle and Fiscal Aggregates GDP and Business Cycle in Kenya The analysis shows that there is a strong seasonality pa ern, with rela vely high produc on in the third and fourth quarters, and rela vely low produc on in the first and second quarters. To iden fy this feature of the series and to make a seasonal adjustment, the SAS version of the US Census Bureau X-12-ARI- MA procedure was used. The seasonally adjusted series is presented in Annex 1: Figure A2. Similarly to the original series, it exhibits an upward trend. To measure a business cycle, we use four different methods of de-trending real GDP: (a) the Piecewise linear filter; (b) the Hodrick-Presco (HP) filter; (c) the Beveridge-Nelson decomposi on; and (d) the Peak-to-Peak trend. Annex 1: Figure A3 illustrates the trends es mated using these four techniques, and Annex 1: Figure A4 presents devia ons of Kenya s real GDP from the trends. All the figures men- oned below are to be found in Appendix Group A. Annex 1: Figure A1, in this Appendix, presents the real GDP series for the period of the study. As can be seen in Annex 1: Figure A4, the cyclical components in three out of four cases are very similar. The cyclical component resul ng from the Beveridge-Nelson decomposi on is smaller in size and exhibits somewhat different behavior. However, as Table 3 below illustrates, even this component is posi vely correlated with the other three. Due to these similari es, only the HP method is used for further analysis. Trends and Cycles in Kenya s Fiscal Accounts The analysis starts with the conversion of revenue and expenditure series into real terms, adjus ng for seasonality in case a strong seasonality pa ern is present, and de-trending using the HP, just as was done for real GDP. Because the GDP deflator is not available at a quarterly frequency for the selected period, we use the CPI instead to convert the series into constant 2001 prices. In addi on to performing this procedure for the total revenue and expenditure, we study each revenue and expenditure category separately. Table 4 presents a summary of selected aggregate variables, while Table 5 provides more detailed informa on. Table 3: Correla ons among various measures of the cyclical component of GDP, Q3 Piecewise Linear HP Peak-to-peak BN Trend Trend Trend Decomposi on Piecewise Linear Trend 1 HP trend Peak-to-Peak trend BN Decomposi on Source: World Bank staff es mates 4

16 Fiscal Policies And Ins tu ons For Shared Growth Table 4: Summary budget figures, Budget category As a percentage of GDP Full sample Revenue Full sample Overall Surplus Primary Surplus Revenue Expenditure Interest Primary Expenditure Source: World Bank staff es mates Table 5: Components of public sector revenue and expenditure, Budget category GDP Full sample Revenue Full sample Revenue Import duty Excise duty PAYE Other income tax VAT local VAT imports Other revenue Expenditure Total recurrent expenditure Domes c and foreign interest Pensions, etc Wages and salaries Other recurrent expenditure Total development expenditure Source: World Bank staff es mates 5

17 Fiscal Policies And Ins tu ons For Shared Growth Revenue Unlike the real GDP trend, the trend in total revenue does not slope upwards, remaining instead rela vely flat for the en re study period (see Annex 1: Figure A5a). Such a pa ern can be be er understood once we examine the behavior of its components. Other revenue and excise duty are the only two categories that have rela vely flat trend lines, similar to that of total revenue. Pay As You Earn (PAYE) and Value Added Tax (VAT) revenue, both on imports and domes c products, grow in significance steadily through me, replacing the revenue from import duty and other income tax revenue, which exhibit downward trends, such that the total does not change substan ally on average. The cyclical components of the revenue categories are plo ed against the cyclical component of real GDP in Annex 1: Figure A6. Table 6 below provides the standard devia ons for the cyclical components of revenue and expenditure, and their correla on with the cyclical component of GDP. The sample is further split into two sub-samples: and , where the second period corresponds to the period when the NARC government was in office. All the revenue components are posi vely correlated with the cyclical component of GDP. For two components excise duty and PAYE this rela onship became stronger over me. The series is extremely vola le, much more so than GDP, as can be seen from Table 6. The other revenue category exhibits the highest vola lity (23.8 percent versus 1.5 percent for GDP), probably because its composi on changes over me. Expenditure Annex 1: Figure A7 presents a set of diagrams of the expenditure components and their trends. We consider the total interest due, rather than the foreign interest due, in sub-plot (d) since the foreign interest due has a nega ve entry and, therefore, we cannot study its logarithm and compute the HP trend. Both the domes c and foreign component of the interest due fall over me in real terms. Pensions and other recurrent expenditures are the only two categories that follow an increasing trend, while the trend for wages and salaries is hump-shaped, with a slow down star ng in 2004 (see Annex 1: Figure A7 (f)). The trend for the total development expenditure, on the other hand, has a U-shape, with a trough approximately in 2001 (see Annex 1: Figure A7 (h)). This structural break can also be a ributed to the policy shi associated with the change in government. The total expenditure exhibits a slight posi ve trend, with a slight accelera on around The expenditure components have lower correla on with the cyclical component of GDP than the revenue components. In the case of pension-related expenditures, the correla on coefficient was actually for the years of ; that is, the pension payments during this period Table 6: Cyclical proper es of public sector revenue and expenditure, 1996 Q Q3 Correla on with GDP Standard devia on Percent *Share of GDP (percent) Real GDP Total expenditure Total recurrent expenditure Domes c interest Total interest due Pensions etc Wages and salaries Other recurrent expenditure Total development expenditure Total revenue Import duty Excise duty PAYE Other income tax VAT local VAT imports Other revenue Source: World Bank staff es mates 6

18 Fiscal Policies And Ins tu ons For Shared Growth exhibited counter-cyclical behavior. The vola lity of the expenditure cyclical components is also higher than that of the GDP component. Domes c interest payments, total interest payments, and total development expenditure have the highest vola lity: 24.4, 23.1 and 50.2 per cent, respec vely. Cyclically Adjusted Budget Surplus Computa onal Issues To compute the cyclically adjusted budget surplus, there is need to narrow down the list of variables we want to adjust for. In other words, we must decide which components of the revenue/expenditure are automa c and which are discre onary. To include a component of revenue/expenditure in this adjustment list, there needs to be a strong, a priori, reason to believe it would behave pro-cyclically/countercyclically. In the case of Kenya, we have included all the revenue categories for adjustments. We believe that no expenditure component can be expected to be moving closely with the business cycle, a priori; that is, we classify all the expenditure components as discre onary. Further, for each of the revenue components, we es mate its elas city with respect to the cyclical component of GDP (see Annex 1: Figure A8 for the relevant diagrams). For this purpose, we run a series of linear regressions described by the equa on below, where refers to the cyclical component of the corresponding revenue category and is the cyclical component of the real GDP. The regression es mates are presented in Table 7. Table 7: Es mates of revenue elas ci es Revenue Category Elas city Import Duty 2.13* (1.27) Excise Duty 1.79** (0.86) PAYE 0.80 (0.67) Other Income Tax 0.26 (1.44) VAT Local 2.07** (0.92) VAT Imports 0.57 (0.85) Other Revenue 1.78 (2.26) Source: World Bank staff es mates The results show that the only two revenue categories that have elas city coefficients significantly different from zero at 5 per cent are excise duty and local VAT. Import duty has a P-value of 0.101; all the other elas ci es are not significant at 10 per cent. Therefore, we adjust for the cyclicality of import and excise duty, and local VAT. The adjustment is performed according to the following formula: where is the total revenue, is primary expenditure, are the primary budget surpluses before and a er the adjustment, refers to the import duty revenue, is excise duty, and is the local VAT revenue; stand for the es mates of output elas ci es of,, and Annex 1: Figure A9 presents overall and primary budget balance before and a er the cycle adjustment as a percentage of the HP trend of real GDP. While both adjusted and unadjusted overall balance exhibit a very slight downward trend, there is a no ceable decline in the primary balance over me. While it remained posi ve throughout the late 1990s, it deteriorated in the first years of the 21st century, reaching -4.9 per cent in 2008 Q2. This behavior is primarily a result of an increase of the noninterest government expenditure, combined with no substan al increase in the total revenues. Panel (e) of Figure A9 depicts the cycle adjustment, also as a percentage of the GDP trend. It never reaches 1 per cent in absolute value, though it is s ll somewhat significant. For instance, the cycle adjustment is nega ve for most of the period of 2002 Q Q1, which implies that fiscal policy was more expansionary than what the non-adjusted measure of the budget surplus indicates. The adjustment reaches the value of per cent in 2002 Q3. Characterizing Fiscal Policy Kenya s fiscal policy does not appear to be pro-cyclical. As can be seen from Table 8, the unadjusted primary balance is posi vely correlated with the cyclical component of GDP. This suggests that it is higher during expansions and lower during recessions, which would suggest counter-cyclicality instead. Some of this effect, however, is due to automa c increase in revenues associated with expansions. Once the primary budget balance is cyclically adjusted, the correla on coefficient decreases. It is es mated to be 0.08 for the full sample, which suggests that the adjusted primary balance is almost acyclical. Further, if we consider a subsample of , the correla on coefficient drops to -0.02, indica ng that fiscal policy became 7

19 Fiscal Policies And Ins tu ons For Shared Growth Table 8: Correla on of the budget surplus measures with the cyclical component of GDP Full sample: Shorter sample: Cyclically Not Cyclically Not adjusted adjusted adjusted adjusted Overall balance Primary balance Source: World Bank staff es mates somewhat more pro-cyclical under the new government. S ll, the magnitude of the coefficient is too small to suggest significant pro-cyclicality. This conclusion is supported by the fact that, according to Table 6 above, the expenditure components are not strongly correlated with the cyclical component of GDP; besides, this correla on is weaker than for the components of revenue. We also compute the discre onary primary budget balance and the fiscal impulse. The discre onary balance is computed as follows: where is the primary balance, t and x are the sample averages of revenue and primary expenditure as frac ons of GDP, Y is real GDP, and Y* is the HP trend of real GDP. The discre onary balance is plo ed in Annex 1: Figure A10 against the cyclically adjusted primary balance. The two series are highly correlated (with a correla on coefficient of 0.998). The fiscal impulse is computed as and provides a measure of the stance in fiscal policy. When the fiscal impulse is nega ve, fiscal policy becomes more contrac onary, and vice versa; a posi- ve fiscal impulse indicates more expansionary fiscal policy. The moving average of the fiscal impulse is plo ed against the moving average of the change in the cyclically adjusted primary balance in panel (b) of Figure A10. Again, the two series are strongly correlated; in fact, they are nearly iden cal. Comparing them with the cyclical component of GDP, one can also conclude that Kenya s fiscal policy does not exhibit strong pro-cyclicality. Impact of Fiscal Policy on Real Ac vity: A Small Var Model of Kenya s Economy Model Features In order to assess the effect of random discre onary fiscal policy (fiscal innova ons that are purely unpredictable) on Kenya s economy, we construct and analyze a small VAR model (see Annex 2). The following variables are included in the model: The logarithm of the world price of oil (we use Europe Brent Spot Price FOB, expressed in constant 2001 Ksh per barrel), pot The logarithm of real GDP in the US (in millions of 2000 chained dollars), yut The US Federal Funds rate (percent per year), rut The cyclically adjusted primary budget balance of Kenya (percent of HP trend GDP), δbt The logarithm of Kenya s real GDP (in millions of constant 2001 Ksh), yt The logarithm of the real Kenya-US exchange rate*t. The price of oil is included as one of the variables because Kenya s trade balance and budget revenue closely depend on it, both through transporta on costs and also directly, since Kenya trades petroleum products interna onally. The GDP of the US is included as a proxy for the world demand for Kenya s exports (specifically, Kenya exports tea, coffee, apparel, household goods, etc to the US, UK, and other countries). The US Federal Funds rate is used to account for the state of the world financial market. Finally, real exchange rate is added to account for other shocks to the Kenyan economy. The technical model is presented in Appendix Group A, Annex 3. The results of the es ma on are presented in two forms: plots of the impulse response func ons (Annex 1: Figure A11) and a table with variance decomposi on (Table 9). Impulse response func ons in this context shows the impact of an output shock on the cyclically adjusted primary balance. Panel (b) of Annex 1: Figure A11 illustrates the response of 8

20 Fiscal Policies And Ins tu ons For Shared Growth the cyclically adjusted primary balance to a 1 per cent posi ve output shock. Due to one of the assump ons made, the adjusted fiscal balance only responds contemporaneously to the external variables and, therefore, the first element is zero. The fiscal balance improves for two quarters and then decreases. The ini al posi ve response only reaches 0.4 per cent of GDP, which suggests that Kenya s fiscal policy is not pro-cyclical. Panel (a) of Annex 1: Figure 11(a)) suggests that fiscal shocks have reasonably low persistence, with the effects of the original fiscal shock negligible by quarter 5. At the same me, the output exhibits somewhat higher persistence, with the effects of the original shock las ng for 9 quarters, see Figure 11(d). Finally, panel (c) illustrates that purely unpredictable fiscal shocks have very insignificant effects on output. While fiscal ghtening causes real GDP to contract, the response is smaller than 0.1 per cent in magnitude (in response to a 1percent change in the cyclically adjusted primary balance), which suggests that the component of fiscal policy we focus on in Kenya, the random discre onary policy, has a limited effect on GDP. However, this finding needs to be interpreted with cau on. There are two other fiscal policy components: (i) systemic discre onary policies, which are rou ne responses to changing economic situa ons; and (ii) automa c policies, governed by rules and laws, such as the tax code. A full assessment of the impact of overall fiscal policy on GDP needs to take all three components into account, and further research is needed in this area. Table 9 presents the variance decomposi on of output for the forecast error in the real GDP of Kenya. Again, fiscal shocks do not appear to have an important effect on GDP. The importance of the output shock decreases with the me horizon, while oil price shocks, shocks to the US GDP (and especially shocks to the Federal Funds rate) become more important in the longer term. In fact, in the long term, the US monetary shocks explain over 70 per cent of the variance in Kenya s GDP. This seems to imply that it is Kenya s monetary policy, rather than fiscal policy, that is an important factor in terms of output vola lity. Further, the fact that Kenya s random discre onary fiscal policy has had limited impact on the country s real GDP seems to suggest that a considerable amount of Kenya s government spending either goes towards government consump on or unproduc ve investment. Annex 1: Figure A12 shows historical contribu ons of the fiscal shock component to output vola lity, again sugges ng low impact of fiscal shocks on Kenya s real GDP. Conclusions and Policy Implica ons This study has analyzed the cyclical proper es of Kenya s fiscal policy. First of all, Kenya s fiscal policy does not lean with the wind. While some revenue components adjust automa cally over the business cycle, increasing in booms and falling in recessions, the correla on of the discre onary budget component with the cyclical component of GDP is small. As a result, the cyclically adjusted primary balance appears to be acyclical for the period Interpre ng our finding in the context of the broader literature on this topic, the acyclicality of fiscal policy suggests that Kenya s fiscal policy has been a growth-enhancing factor. As shown by a number of researchers, cyclicality is usually associated with vola lity, which in turn lowers economic growth. The finding that fiscal policy has been acyclical in Kenya would imply that fiscal policy did not amplify extraneous shocks during the study period, and this may have supported growth. However, it is worthwhile emphasizing that fiscal shocks by themselves are o en cause of macroeconomic instabili es, and a major fiscal shock would probably have led to macroeconomic instability in Kenya. Second, the random discre onary component of fiscal policy of Kenya does not appear to have a substan al effect on GDP, though this finding requires a careful interpreta on. While fiscal ghtening has a contrac onary effect, this effect is very small in magnitude and rela vely short-lived. However, the other components of fiscal policy, namely the systemic discre onary policy and automa c policies, need to be also considered to assess the impact of the overall impact of fiscal policy on GDP in Kenya, which is a topic for further research. The results of the VAR analysis corroborate the finding that fiscal policy is acyclical in Kenya. In par cular, the variance decomposi on analysis suggests that there is an important rela onship between the US Federal Funds rate and the cyclical component of Kenya s GDP. Therefore, Kenya s monetary policy, and specifically its exchange rate policy, appears to be an important determinant of Kenya s GDP fluctua ons. Which factors may have contributed to 9

21 Fiscal Policies And Ins tu ons For Shared Growth these findings? One possibility is the fiscal adjustment Kenya experienced following the Goldenberg scandal. Expenditures between 2002 and 2007 remained fairly steady and ranged from per cent, and these series do not exhibit much vola lity. Public borrowing was limited to about 1.8 per cent of GDP in that period, and the fact that there were no major differences between policy announcements and fiscal outcomes would suggest that fiscal policy reduced uncertainty in general. This may have contributed to this finding, which suggests that Kenya should avoid major changes to its fiscal policy. Although somewhat surprising, the results of this study should be interpreted with cau on. It is well known that the quality of any data analysis is determined to a great extent, by the quality of underlying data. In this respect, there are two main factors to consider: the general reliability of the data, and the somewhat short me horizon. Without doubt, it would be beneficial to re-examine the issue once more data becomes available. Table 9: Variance decomposi on of output (percent of variance of the forecast error) Forecast Percent of variance due to shocks to Horizon Oil price US GDP Federal Fiscal GDP Real exchange (quarter) funds rate balance rate Source: World Bank staff es mates 11

22 2. A GLOBALLY COMPETITIVE AND PROSPEROUS KENYA: THE ROLE OF PUBLIC INVESTMENT Introduc on This note explores the effec veness of public investment in Kenya and makes recommenda ons for improvements. It is the second in a series of budget policy notes exploring fiscal policy, growth, and poverty linkages. The note is based on a review of the literature and publicly available fiscal data. It a empts to examine the role of public investment as a policy tool for: (a) achieving the aspira- ons set out in the government s Vision 2030; (b) achieving broad-based economic growth in Kenya; and (c) improving service delivery for household welfare improvements. The note focuses on the ins tu onal issues that have adversely affected the performance of Kenya s development budget, such as the procurement procedures and the donor disbursement delays, which have resulted in significant under-spending in recent years. In light of the 2009/10 fiscal s mulus package to bolster development spending, expenditure under the development budget is planned to grow from 6 per cent of GDP in 2007/08 to an average of 8 per cent of GDP per annum in the next five years. This note also provides some conclusions and recommenda ons for improving the use of the development budget for sustainable development outcomes. While opera ng and maintaining public investments is cri cal for the full benefit of the ini al investment to be realized, a full analysis of the adequacy of the recurrent expenditure goes beyond the remit of the policy note, and provides an area for follow-up work that could be taken forward in the forthcoming Public Expenditure Review (2010). The note starts with an overview of public investment outcomes and the state of the physical capital stock in Kenya. The note also analyzes the composi- on and trends in development budget spending, and examines the budget process issues and the extent of proposals for improvements in the short and medium term. Public Investment: Why It Ma ers There are three core reasons why public investment ma ers. First, public investment provides goods and services to households, which directly improves their welfare. For example, public investments can result in households gaining access to electricity or piped water and sanita on, resul ng in vastly improved health status and quality of life. Second, public investment is posi vely correlated with a reduc on in poverty o en through more indirect means. For example, investments in rural roads facilitate farmers ge ng their products to new and wider markets, or children to school. Third, public investment is also empirically posi vely correlated with economic growth through lowering the costs of doing business and crowding in private investment. For example, cheaper and more reliable access to electricity, road and rail infrastructure lowers the costs of doing business and creates new investment opportuni es for future economic growth. However, public investment can also crowd out private investment, par cularly if it is financed domes cally, causing interest rates to rise and increasing the cost to the private sector of undertaking its own investments. 12

23 A Globally Compe ve And Prosperous Kenya Box 1: Public investment: Does it crowd in or crowd out private investment The role of the public sector in determining private investment and economic growth is covered in several empirical studies. Da ng back as far as 1989, Aschauer found that public investment crowds in private investment spending in both developed and developing countries.² There has been a resurgence of empirical research into the rela onship as a result of endogenous growth theory, which predicts that the output elas city to private capital is considerably higher than predicted by the private capital share in na onal income. The posi ve impact of public investment spending found in both private investment and growth equa ons has been a ributed to the presence of posi ve spillover effect generated from the provision of quasi-public goods, whose services are essen al for the proper func oning of a market economy. It is argued that public investment by the state in roads, airports, ports, energy, water, State of Infrastructure The stock of capital and access to related services in Kenya lags behind the middle income countries to which it aspires, and is closer to the average in Africa. Kenya s indicators of public investment outcomes lag considerably when compared to a sample of middle-income countries with respect to access to sanita on, telephone, internet and paved roads. Kenya is closer to LICs and SSA averages (see Table 10 below). Kenya should not feel comfortable with being average; on just about every measure of infrastructure coverage, the Africa region lags other health facili es and educa on services tend to complement private capital forma on by eliminating bo lenecks. However, if such capital spending is financed by domes c borrowing, this pushes up the cost of capital and reduces private investments accordingly. Poorly designed public investment projects and/or those with low or marginal rates of return would, therefore, be not only poor value for public money but could poten ally have a doubly nega ve effect through the impact on domes c interest rates. Crowding out can also occur with other inputs to investment projects. For example, the costs in the construc on industry with a limited supply of contractors and professional services can increase drama cally, or even be totally absorbed by the public investment program, raising the costs of such services to the private sector and making those investments less viable. parts of the developing world. The differences are par cularly large in the case of paved roads, communica ons infrastructure and power genera on capacity. Furthermore, for these three key infrastructure sectors, Africa has been expanding stocks much more slowly than other developing regions, and therefore the gap is widening.³ Even if the medium-term targets for expected outcomes on many of the infrastructure indicators are met, this would s ll place Kenya in 2011 at or below today s LIC averages. Table 10: Public investment outcome indicators, Kenya in comparison, 2008 Kenya Sub-Saharan Africa LIC average Other Low Income Country average Paved road density (kilometers per kilometer squared) Total road density (kilometers per kilometer squared) Main line density (lines per thousand popula on) Mobile density (lines per thousand popula on) Internet users (subscribers per thousand popula on) Genera on capacity (megawa s per million popula on) Electricity coverage (percent of popula on covered) Improved water (percent of popula on covered) Source: Yepes et al. (2008), World Development Indicators (2008) See also Cardoso (1993), Ram (1986). Foster (2008). 13

24 A Globally Compe ve And Prosperous Kenya In the case of access to electricity, Kenya lags behind LIC averages by a considerable margin, and is on a par with low income SSA averages. Although access in Kenya (18percent of households) is ahead of neighboring Uganda (9percent have access to electricity) and Tanzania (11percent have access to electricity), it lags behind the average for the low income countries at 41 per cent. Furthermore, the 2011 target to connect 33 per cent of households to electricity is s ll below the current average of 41 per cent with electricity in LICs in 2008 (see Table 10). Africa s largest infrastructure deficit is in the power sector in terms of genera on capacity, electricity consump on, and security of supply. Power consump on at 124 kilowa hours per capita per year and falling is only a tenth of that found elsewhere in the developing world. The situa on for roads, piped water and sanita on is closer to the averages for LIC and SSA. Kenya has a rela vely well extended road network, which is 112 road miles per km2, close to the SSA average of 137 road miles per km2. However, only 14 per cent of the road network is paved. This infrastructure bo leneck is a significant constraint to growth and compe veness. In investment climate assessments, the private sector has consistently iden fied poor and costly infrastructure as the binding constraint to business compe veness in Kenya.⁴ Around half of businesses surveyed in 2007 found infrastructure services (transport and electricity) to be a major obstacle, and this was higher than in the previous survey in Electricity has become more of a problem for businesses over the last four years. As the rate of economic growth during increased, so did electricity demand and therefore the reliability and cost of power supply worsen. Power outages have increased from an average of 16.4 hours per month to 33 hours per month between 2002 and According to the firm surveys, 85 per cent report experiencing power outages, and close to 80 per cent experience losses as a result. Two thirds of firms have their own generator to help resolve the issue. A generator is costly to run and the capital investment of a generator accounts for approximately 3 to 5 percent of the total value of machinery and equipment. The investment climate surveys show that Kenyan firms suffer much higher power costs (indirect costs and losses are about 7percent of sales) than firms report in China, India and South Africa. Transporta on is the other top complaint of Kenyan firms as a result of high associated direct and indirect costs. Unsurprisingly, firms outside Nairobi perceived this as a major problem more o en than firms located in the capital city. Firms in the manufacturing sector are par cularly affected by the poor state of the transporta on system. Inland transport costs in Kenya are much higher than in China and India. Indirect costs are also higher than those of comparators, since Kenyan companies lose 2.6 per cent of their sales to spoilage and the during transporta on. Half of the transporta on losses are due to transport delays, while the other half are due to the during transporta on. The poor suffer more from lack of access to electricity, water and sanita on. About half of the popula- on does not have access to water and sanita on services, and do not therefore meet the minimum standards of quality; the per cent of those meeting the minimum standards is much lower in rural areas. In both rural and urban areas, the poorest households have the lowest access to private piped water supply. On average, it takes 50 minutes in rural areas and 20 minutes in urban areas to collect water. There are also significant regional varia ons in infrastructure access. In Western Province, for example, 6 per cent use piped water and only 4 per cent of the popula on has access to electricity. This is not only a rural phenomenon; the higher urban averages hide a significant rich: poor divide in terms of infrastructure access. In the urban slums, less than 6 per cent have access to piped water, and less than 3 per cent have access to a private latrine. There has, however, been improvement in access to infrastructure over the last decade (see Figure 9 and 10 ). The average distance to a motorable road, for example, has fallen for both rich and poor households. However, the pace of electrifica on has been slow; an average 0.5 per cent of the popula on gaining access each year since While elec- ⁴ World Bank Investment Climate Assessment 2002 and

25 A Globally Compe ve And Prosperous Kenya trifica on across the country increased by 50 per cent between 1993 and 2003, it has since slowed down and access remains low especially in rural areas. Na onally, access to safe water stagnated un l 2003, and actually fell in Western Province during this me. Stock of Human Capital On several indicators of human capital stock, Kenya appears to be no more than average for the SSA region and LICs. A broader concept of public investment should include investment in human capital. Figure 9: Provincial distribu on of service gains, /06 Before looking at the sectoral composi on of the development budget in more detail in the next sec- on, we look at the trend and comparison of Kenya s indicators of human capital stocks. We find that indicators of the health status, level of educa on and employability of Kenyans to be around the SSA and LIC averages (see Table 11 below). However, once again, we find that Kenya lies behind several African countries (such as Ghana) and the Asian middle-income countries that we have used for comparison. The full table of indicators is summarized in Appendix Group B, Annex 2 Table B2. Figure 10: Reduced distances to various infrastructure, Source: World Bank (2008b) Source: Nyoro etal 2008 Table 11: Indicators of human capital, Kenya and SSA and LIC averages Kenya Sub-Saharan Africa average LIC average Employment to popula on ra o, 15+, total (percent) Employment to popula on ra o, ages 15-24, total (percent) Primary educa on, dura on (years) Immuniza on, DPT (percent of children ages months) Immuniza on, measles (percent of children ages months) Life expectancy at birth, female (years) Life expectancy at birth, male (years) Life expectancy at birth, total (years) Mortality rate, infant (per 1000 live births) Newborns protected against tetanus (percent) Source: World Development Indicators (2008) 15

26 A Globally Compe ve And Prosperous Kenya Overview of the Development Budget A Note on Classifica on Public investment is the public sector spending that typically adds to the capital stock; it is approximated by the development budget in Kenya. Public investment is capital spending or acquisi on of non-financial assets in the budget according to Government Finance Sta s cs (GFS 1986 and 2001) Manual classifica on. No such classifica on exists in the Kenya finance sta s cs. Instead, public investment spending is approximated by spending under the development budget. In any par cular year, the non- capital part of the development budget can be sizeable. For example, the 2007/08 development budget es mates include provision for Ksh 41.5 billion, or 20 per cent of the development budget, as the government s equity par cipa- on in the na onal telecom company prior to priva za on. There is some misclassifica on, but possibly less than other countries in East Africa. The development budget comprises all donor-financed spending, regardless of whether these are recurrent costs or capital investments, plus capital projects financed by the government s own sources, Cons tuency Development Funds, and transfers to other parts of the public sector. The donor-financed projects can also contain a considerable amount of funding for recurrent expenditures, such as for goods and services, such as consultants and pharmaceu cals in health projects. However, an exercise to compile public investment spending in infrastructure found that less than 5 per cent of current expenditures were misclassified as development, and only approximately 1 per cent of capital spending was misclassified as recurrent. The sum of misclassifica on problems in the infrastructure sector was less than other countries in the region.⁵ The development budget is predominantly, but not only, capital spending. The economic composi on of the development budget shows that acquisi on of non-financial assets and capital grants are the largest components (see Table 12). Compensa on of employees are less than one per cent on average, and goods and services consump on averages 10 per cent of the budget. In 2008/09, spending on the acquisi on of non-financial assets was largely from the development budget (89percent). However, the development budget includes other types of spending, and only 63 per cent of the development budget was on acquisi on of non-financial assets in 2008/09, and this was as low as 41 per cent of the development budget the year before.⁶ There is poten ally a considerable amount of public investment spending that takes place off-budget. Briceno-Garmendia and Foster (2007) take a more comprehensive look at infrastructure expenditures by looking at the infrastructure spending of several non-budget ins tu ons.⁷ They find that only around 30 per cent of infrastructure spending in Kenya is on budget. This varies according to sector, from less than 10 per cent of energy expenditures to almost 70 per cent of transport expenditures. Given that about half of infrastructure spending is public investment, and many of the non-budget ins tu ons are state-owned enterprises, this could imply that considerable off-budget public investment is taking place. It is not surprising that a substan al share of public investment is off budget and nor is it necessarily sub-op mal. Public investment in infrastructure is o en carried out through the relevant state-owned enterprises, rather than the central government. This is par cularly the case when such investments can be recouped through the sale of services to the private sector, but may also require transfers from the budget and implicit and explicit loan guarantees when the capital for those investments is raised. Most of the pressing demands for infrastructure in Kenya such as electricity are provided by parastatals. Therefore, analysis of the infrastructure that public enterprises finance and the scope for full cost recovery, including the servicing of their capital, should be undertaken to more accurately ⁵ Briceno-Garmendia and Foster (2007). ⁶ Government of Kenya (2009). It will be useful in follow-up work to explore further the use of capital grants and transfers included in the development budget. 16

27 A Globally Compe ve And Prosperous Kenya Table 12: Economic composi on of development budget, 2004/ /09 as percent of total 2004/ / / / /09 Compensa on of employees Use of goods and services Subsidies and grants Capital grants Acquisi on of non-financial assets (net) Others Source: Government of Kenya (2009) assess the full poten al for Kenya to raise the funds required for infrastructure investments. This analysis was not undertaken here due to the difficulty of accessing the relevant data.⁸ Previous World Bank research on infrastructure in East Africa indicated that u li es in Kenya incurred substan al quasi-fiscal deficits, resul ng in insufficient investment and maintenance, which results in a deteriora ng capital stock. Further analysis of this should be taken forward in the Public Expenditure Review Addi onal work is also being done separately on the role of the private sector in public investment. Increasing Public Investment Kenya The Kenyan development budget has risen from 3.0 per cent of GDP in 2003/04 to 6.6 per cent of GDP in 2007/08, and an es mated 9.3 per cent of GDP in 2008/09. It is forecast to stay at approximately 8.0 per cent of GDP over the medium-term. This increase in development spending has been brought about by fiscal space created by an increase in revenue from 21 percent of GDP in 2003/04 to 24.4 per cent of GDP in 2008/09, and restraint on the growth of recurrent expenditure (maintained at an average of 20 percent of GDP over the period). The forecast shows that a combina on of increased domes c and external borrowing will be required to accommodate a higher level of development budget in propor on to GDP. Table 13 shows the evolu on of the fiscal outcomes since the beginning of the decade and the medium term plan. In real terms, the development budget has increased almost four-fold since the beginning of the decade. Figure 10 shows that total development spending has increased from Ksh 25 billion to Ksh 97 billion between 2000/01 and 2008/09 in constant prices. Between 2006/07 and 2007/08, there was a real increase of 50 per cent in public investment, and there is an es mated further 28 per cent increase in 2008/09. While Kenya has reversed the decline in public investment that occurred in the 1990s, it has yet to reach the historically high rates of investment seen in middle-income country comparators. Using the IMF s GFS01 fiscal dataset, we compared Kenya s spending on non-financial (mostly fixed) assets over me with several comparators (see Figure 11 ). We find that as a percentage of GDP, there was a considerable decline in public investment spending over the 1990s, a trend that was reversed in the following decade. However, at less than 4 per cent of GDP in 2007/08, this is s ll below the investment spending as a percentage of GDP of Indonesia and Malaysia in the 1990s, and South Africa s average since As richer and larger economies (in terms of GDP), this also means that in per person terms, this is considerably lower. ⁷ These include the power u li es and several private and state-owned enterprises providing the rail, airport, and telecom services. ⁸ Comprehensive details for the 161 state corpora ons on their expenditure and non-tax revenues are not included in the fiscal reports, either in consolida on with other central government expenditure, or shown in a separate annex of the document, or shown in a separate document presented to the legislature and published at the same me as the fiscal reports. 17

28 A Globally Compe ve And Prosperous Kenya Table 13: Fiscal space for increased development spending 2002/ /09 Revenue Expenditure Recurrent Development Net lending Balance Financing Of which 2002/ / / / / / / / / /12 Domes c External Source: Government of Kenya, Budget Strategy Paper 2009; Quarterly Economic Budget Review Third Quarter 2008/09; IMF Staff Report for the 2008 Ar cle IV Notes: Data from 2002/03 to 2007/08 are actuals, data for 2008/09 are es mates, and data for 2009/10 to 2011/12 are projec ons Figure 11: Rise in development spending, 1999/ /12 Manpower and ICT Health Jus ce, Law & Physical Infrastructure / / / / /09 Source: IMF and MoF Notes: Constant 2001/2 prices 18

29 A Globally Compe ve And Prosperous Kenya While Kenya s total spending on infrastructure is high compared to East Africa (10 percent of GDP), only 20 per cent is for public investment. Given the limita ons of GFS-classified data with respect to iden fying infrastructure spending, a new fiscal dataset for Africa was compiled for an earlier World Bank report. Their findings show that there is rela vely low public investment in infrastructure in Kenya (only 2percent of GDP). While spending on infrastructure in propor on to GDP is around 10 per cent in Kenya the highest in the region and slightly higher than middle income countries in other regions the low share that goes to public investment means that Kenya is much lower than Uganda and Rwanda, where over half of infrastructure spending is on public investment. Low public investment compared to the region is confirmed by the government s Public Expenditure Review (2009). The Review reports that Kenya has been behind other countries in the East African Community on capital investment. Kenya s development budget has been rela vely low compared to Tanzania and Uganda. In 2004/05 and 2005/06, Tanzania and Uganda were inves ng more than double the level in Kenya in propor on to GDP. Table 14 shows how Kenya compares, and the progress that has been made since then as development spending increases. Priori es are Shi ing Towards Physical Infrastructure The composi on of development spending shows a shi towards physical infrastructure in recent years. The sectoral composi on of Kenya s development spending as reported in Government of Kenya (2009) shows that the shares have shi ed (see Figure 12 and Table 15) to reflect an increase in physical infrastructure, largely at the expense of shares to health, educa on and governance sectors. Taking a longer me perspec ve, and comparing the averages for two periods, pre- and post- 2003, both physical infrastructure and public administra- on made the biggest increases in the share of the development budget. As shown in Figure 13, physical infrastructure averaged less than 30 per cent of development spending in the period, and the share increased to 39 per cent of spending for the period Educa on also gained from 6.1 per cent to 7.8 per cent of spending. The biggest increase, however, was in public administra- on, which doubled its share of spending from 16.5 per cent to one third of development spending. This reflects, in part, the increase in development budget funds going through the Ministry of Finance, including for equity investments as men oned in the notes on classifica on sec on. The sector shares for agriculture and rural development and public safe- Figure 12: Acquisi on of non-financial assets, Kenya and comparators, Source: GFSO1 Briceno-Garmendia and Foster (2007). 19

30 A Globally Compe ve And Prosperous Kenya Table 14: Kenya and comparators, development spending as percent of GDP Tanzania Uganda Kenya 2004/ / / Source: Government of Kenya (2009) Figure 13: Share of development spending by sector, per cent of total 2004/ /09 Source: Economic Survey 2009; East African Facts ans Figures, 2006 Table 15: Share of development budget by sector, 2004/05 to 2008/09 (Ksh millions) 2004/ / / / /09 Produc ve Sector 19, , , , , Public Administra on 26, , , , , Physical Infrastructure 34, , , , , Governance, Jus ce, Law and Order 47, , , , , Health 19, , , , , Educa on 81, , , , , Na onal Security 25, , , , , Informa on & Communica on Technology 1, , , , , Manpower and Special Programs 2, , , , , Total 257, , , , , Source: Government of Kenya Public Expenditure Review 2009 ¹⁰ Although the infrastructure deficit is well documented, it would be interes ng for future work to look at the composi on of the infrastructure investment, since the Foster (2008) found that Kenya (like others in the region) showed signs of under-investment in the power sector, and poten al over investment in the water sector. 20

31 A Globally Compe ve And Prosperous Kenya ty and law and order declined significantly. While increasing in real Kenya shilling terms, the share of net development spending in the health sector fell to 5 per cent, on average. In the last two years, the budget shows a slight shi back to priori zing the social sectors or human resource development as the sector has been defined in the Medium Term Expenditure Framework (MTEF). In real terms, there has been a decline in resources to the social sectors. Reclassifying the agency or ins tu onal breakdown into standard sectors and adjus ng to constant prices shows that not all sectors have seen a real increase in development budget spending. The increase in spending on physical infrastructure has been maintained, and so too the rise in public administra on development spending. Health and educa on, however, have been declining in real terms in the most recent years. Role of Development Partners The overall role of development partners in financing the development budget is decreasing. Annual development partner grant expenditures rose from Ksh 4.4 billion to Ksh 33.8 billion between 2002 and 2008, and loan disbursements from Ksh 14 billion to Ksh 47 billion (see Table 17). However, the donors are no longer financing the majority of the development budget, falling from 73 per cent of expenditure to 41 per cent of the development budget in 2008/09. In terms of annual commitments to the Government of Kenya, the largest development partner is currently the World Bank (officially the Interna onal Development Agency-IDA). Figure 14 shows the loan/grant mix of donor assistance as captured by External Resources Department (ERD) data on donor commitments. The top two donors: IDA and the African Development Bank made up 43 per cent of total donor commitments in 2008/09, followed by France, China and the EC, who each have around 6 per cent of total annual commitments. Development partner finance is predominantly for the infrastructure ministries. Figure15 shows the main government counterparts listed in the 2008/ 09 budget as the main recipient for donor funds. As we might expect, the ministries responsible for physical infrastructure (Ministry of Roads, Ministry of Water and Irriga on and Ministry of Energy) are the most significant beneficiaries and implemen ng agencies for development partner-funded projects and programs. Figure 14 : Ministerial Spending 2000 Prices: Spending shi to physical infrastructure Source: MoF Key: Agric.and R. Dev- Agriculture and Rural Development; P. Infrast. - Physical Infrastructure; PBSLO- Public Safety Law and Order; P. Admin.-Public administra on. 21

32 A Globally Compe ve And Prosperous Kenya Table 16: Sectoral composi on of development expenditures (Ksh millions) 2004/ / / /08 Public Administra on 18,496 15,191 30,280 57,555 Physical Infrastructure 15,853 22,543 35,512 33,517 Agriculture and Rural Development 1,652 3,629 5,409 7,021 Educa on 3,554 5,846 5,492 4,536 General Economic Services 1, ,794 4,098 Public Safety Law and Order 5,017 4,399 5,659 3,704 Health 595 2,939 2,163 2,748 Total Development 46,463 55,508 89, ,179 Source: Ministry of Finance Notes: Constant 2004 prices Table 17: Donor-financed development expenditures, 2002/ /09, Ksh billions 1999/ / / / / / / / / /09 Grants Projects Programs Loans Projects Programs Total Source: Ministry of Finance, Kenya Figure 15: Total donor commitments 2008/09 Source: World Bank staff es mates 22

33 A Globally Compe ve And Prosperous Kenya Figure 16: Government departments with donor-financed projects, 2008/09 budget Key: MOR-Ministry of Roads; MOWI-Ministry of Water and Irriga on; MOE-Ministry of Energy; MOPBHS-Ministry of Public Health and Sanita on; MOA-Ministry of Agriculture; MOEDU-Ministry of Educa on; MOF-Ministry of Finance; CAB-Cabinet Office; MOSSP-Ministry of Special Programs; MOPND-Ministry of Planning and Na onal Development and Vision2030; MOT-Ministry of Transport. Development Budget Performance Absorp on Capacity Problems The development budget has suffered from poor execu on rates in the past, raising concerns that the budgeted increase in spending would not translate into outcomes due to absorp on constraints. Public investment is o en for projects that are implemented over a number of years and, therefore, require a medium-term approach for planning and public finance alloca ons. This makes these investments difficult to accommodate in a single budget cycle, and raises the risks that mul -year investment projects are not sustained over me, or may not receive sufficient annual alloca ons to keep the project on schedule. The annual cycle does not lend itself to planning and procuring for public investment projects, which consequently results in a low execu on of the budget and eventually leads to severe delays in implementa on and stalled projects. Even in highly developed budget systems, there is a tendency to under-spend on the capital program compared to recurrent or opera onal budgets. Projects may experience delays for a variety of reasons (such as weather condi ons or poli cal disrup- ons) that lie beyond the control of their managers. A well managed public spending program would allow budget managers to switch resources from stalled projects to those where implementa on can be accelerated, thereby par ally offse ng the impact on the overall capital spending program of such delays. In countries where a substan al share of public investment comprises externally financed projects, levels of under-spending have tended to be significantly higher due to both the complexity of development partner procedures and lack of flexibility to switch resources between projects when delays occur. Over-op mis c and unrealis c budge ng are also frequent problems. The budget process could be improved to ensure that high return projects are priori zed and sustainable. The development budget is developed at the same me as the recurrent budget, according to the standard budget cycle from 1 July to 30 June. The development budget is prepared separately with no linkages to the recurrent budget. The budget cycle is captured in Figure 16 and the key players and stages in the budget cycle are explained in Appendix Group B, Annex 1. 23

34 A Globally Compe ve And Prosperous Kenya Box 2: Public investment management: A note on good prac ce Drawing on World Bank country case studies on public investment management (PIM) and the development of a new framework for analysis, we put forward the following stylized rules for good public investment management prac ce: Clear centrally-issued guidelines: Guidelines for the development of public investment projects, which are then rigorously applied, are a cri cal aspect of a well func oning PIM system. Provide training: There is need to develop pragma c systems based on how the rest of the budget management works, and to ensure adequate training for prac oners in their use, and the poli cal will to enforce the system. Evidence based research: Availability of good data is a cri cal component in the evalua on of the efficiency of specific public investment projects, and for reviewing the performance of PIM systems as a whole. Some countries, most notably Chile and Korea, are able to assemble and publish a great deal of investment-related data. Role of Public-Private Partnerships (PPPs): There is general acceptance that PPPs can, for a number of reasons, accelerate the delivery of highly valued social services and infrastructure. The availability of PPP-related financing and use may need to be considered in the context of the availability of relevant exper se to assess PPP arrangements, the poten al for op mal risk transfer, and value for money considera- ons. The Australian case usefully highlighted a number of technical issues associated with managing a PPP por olio. To conclude there is general agreement that effec ve PIM needs to be supported by the assembly of good data, access to well-trained officials, and the need for support of government at the highest level. Source: PIM Conference Seoul, Korea, 21 November

35 A Globally Compe ve And Prosperous Kenya Figure 16: The budget process 1. Ministry of Finance ini ates budget cycle 2. Line ministries prepare budget submissions SWGs 6. Parliament discusses and approves budget 3. SWGs decide sector priori es and provide recommenda ons to the EABC. The EABC reviews overall budget ceilings and dra s BSP 5. Ministry of Finance finalizes budget es mates prior to submission to Parliament 4. EBSC prepares BSP and oversees sectoral budgets 25

36 A Globally Compe ve And Prosperous Kenya The main deficiencies in the budget-approval cycle that relate to the performance of the development budget are: Development projects are not priori zed according to economic appraisal, ment budget, projects that require more than one year to be implemented are not required to develop medium-term financing implica ons (neither capital alone, nor capital and recurrent) consistent with the MTEF and outer year ceilings. Development projects are not clearly iden fiable in the development budget es mates or government accounts, hampering monitoring of development project progress. While the Sector Working Groups (SWGs) no onally apply a sectoral priori za on filter for development projects within a sector, the direct nego- a ons that take place later in the cycle serve to negate the relevance of this step. The budget is highly contestable, and line ministries in addi on to direct nego a on also have a publicity campaign to increase the importance of their programs. The final decision on alloca ons rests with the Ministry of Finance, and this has also led to cuts in the development budget for projects that the line ministry would not have themselves cut. On average, one third of the development budget is not spent each year. As Table 18 shows, the execu on rate of the development budget has been improving from around 40 per cent of the budget in 2001/02 to over 70 per cent by 2006/07. However, budget execu on rates vary significantly by sector and by year. The execu on rate in the health sector is especially low, and only 22.4 per cent of the budgeted development expenditures was spent in 2006/07. Despite the a empts to improve execu- on rates, there remain significant problems of under-performance in the development budget. Projects take too long to finish and ci zens fail to benefit from the spending on me. Table 18: Development budget execu on rates, 1999/ / / / / / / / / / 2008 Agriculture and Rural Development Public Safety, Law and Order Health Educa on Physical Infrastructure General Economic Services Public Administra on Total Development Budget Total All Budget n/a Source: Ministry of Finance (Kenya), as reported in the Concept Note 2007/08 Bank staff es mates Notes: 1- n/a indicates data unavailable., 2- Figures for 2003/04 Missing 26

37 A Globally Compe ve And Prosperous Kenya There is significant devia on between budgets and spending (even excluding donor financing) at the level of the spending agency, but this is improving. The 2008 Public Expenditure and Financial Assessment (PEFA) provides informa on on the performance of the development budget financed by development partners. Across the board, there was an average devia on of 13.2 per cent between development budget and spending in 2006/07. This is down from 36 per cent in 2005/06. There is considerable varia on, but there are also improvements at the level of agencies, par cularly in infrastructure ministries such as energy, roads, transport, water and irriga- on. With some priori zed effort, the total devia on between government development budgets and spending could be further improved. Only six ministries account for three quarters of the development budget devia on. Not all government ministries have an equal share of the development budget, and therefore some devia ons ma er more than others for the aggregate budget performance. As Figure 17 shows, as a share of the total extent of development budget devia on, there are six ministries that warrant special a en on: Ministry of Finance, Ministry of Roads, Office of the President, Ministry of Defense, Ministry of Special Programs and Ministry of Energy, which together comprise almost 75 per cent of the annual development budget devia on. Figure 18 shows the top under-spenders in Progress has been made in comple ng stalled projects, which might explain the improvements in execu on rates. A er the adop on of the MTEF, the government s policy was to not only increase development spending, but also to increase the comple- on rates in programs and projects. A special alloca on of Ksh 2 billion was specifically allocated for stalled projects each year from 2003/04, and was es mated to take a total of Ksh 8 billion. A total of Ksh 6 billion has been budgeted over the period 2004/ /08, but only about 63 per cent of funds were absorbed. Although execu on rates for these projects was below 50 per cent the first year, it increased to 100 per cent in 2007/08, and a total of 130 projects had been completed by the end of that year. This might explain the improved execu- on rates observed in Table 18.¹¹ Donor disbursements differ drama cally from what is budgeted, pulling down development budget execu on rates. This is examined in more detail in the next sec on, which deals specifically with execu- on problems. Although there are problems with disbursements, there are also serious problems in the repor ng, which makes assessing the extent of execu on very challenging. In the Public Expenditure and Finance Assessment (PEFA) 2008, the indicators for financial informa on provided by donors scored a D+ and concluded that there is insufficient repor ng by donors for budget es mates and accoun ng. At least half of the donors provide budget es mates using their own classifica on, which are not consistent with the government s budget classifica on. Donors provide quarterly reports within two months of end-of quarter disbursements for at least 50 per cent of the externally financed projects in the budget, but less than 50 per cent of aid funds to government are managed through na onal procedures. Part of the problem is over-budge ng for expected donor disbursements. In fact, there are several repor ng differences that confuse the calcula on of donor-financed project execu on rates. According to the Survey on the Paris Declara on in 2008, only about 60 per cent of what had been recorded in the budget as development partner expenditures was due to be disbursed according to the development partner s own records. Of this, as Table 19 shows, there was not only a significant varia on between budgets and disbursements, but there was also a large degree of varia on between what enters the government s budget records and what donors believe they will disburse, and also between what the government has accounted for as having received, and what the donor accounts for as disbursements. The extent of the problem by donor is summarized in Table 19. Alignment with Na onal Goals While Kenya has a well established and ins tu onalized process of drawing up plans to implement a na onal agenda, the process for aligning spending with these plans is not so well established. Vision 2030, the government s new long-term development plan, aims to transform Kenya into a middle-income, industrialized country by To ¹¹ Stalled projects are mainly construc on of office blocks, staff houses, health facili es and police sta ons. 27

38 A Globally Compe ve And Prosperous Kenya achieve this vision, a target of 10 per cent annual GDP growth will need to be sustained. Under the economic pillar of Vision 2030, six key focus sectors are iden fied to drive these annual growth targets: (a) tourism, (b) agriculture, (c) manufacturing, (d) wholesale and retail trade, (e) business process outsourcing, and (f) financial services. In the social pillars, the focus is on: (a) educa on and training (b) health (c) water and sanita on, (d) environment, (e) housing and urbaniza on and, (f) gender, youth and vulnerable groups. The Vision is to be implemented through a series of five-year medium term plans. There are various checks during the annual budget cycle that budgets are aligned with sector and na onal plans, but this is not well-ins tu onalized, and the direct nego a ng between Ministry of Finance and line ministries during the process could serve to undermine the a empts made at sector level to align spending to priori es. An es mated Ksh 500 billion is needed for investments over the next five years. The medium-term plan (MTP) for captures the policies and flagship projects by which the goals of Vision 2030 are to be achieved. In order to implement the flagship projects, the government es mates that Ksh 500 billion will need to be invested, of which the government will invest 50 per cent over the next five years. The government expects the other 50 per cent of the finance to come from local and foreign private investors, mainly through Public- Private Partnership (PPP) arrangements. The MTP iden fies priori es and ac ons that will require public financial commitments and these are summarized in Box 4. Figure 17: Budget devia on by ministry Source: World Bank staff es mates Figure 18: Top 5 under-spenders Source: World Bank staff es mates 28

39 A Globally Compe ve And Prosperous Kenya Box 3: Vision 2030 and the New Coali on Government Vision 2030 was first dra ed prior to the 2007 General Elec on. Following the disputed outcome of the General Elec ons in December 2007, the poli cal and socio-economic environment changed significantly. Although Vision 2030 was revised in early 2008 and has been adopted by the new coali on government, the revisions arguably did not take sufficient considera on of the impact and consequences of the post-elec on violence. Both tourism and agriculture have suffered and will require significant efforts to recover back to 2007 status. Physical damage to public and private property, and the of public property (the energy sector experienced losses due to the the and destruc on of transformers and other expensive equipment) are all near-term priori es that are likely to have budgetary implica ons. Instead of star ng the Vision 2030 period with growth rates of 7 per cent or more (as in 2007), the growth rate in 2008 was only 1.3 per cent. Furthermore, the 2008 downturn in the global economic environment poses a risk to the strategy as the tourism industry globally has taken a down turn, and poten ally through lower FDI and/or remi ances. So far, neither of these impacts have been significant, but the risks remain. Vision 2030 has not, therefore, got off to the promising start envisaged at the me of dra ing. Furthermore, the poli cal situa on remains fragile, with a number of issues (e.g. item four agenda of the na onal peace accord) requiring the a en on of poli cians and public servants over the next five years before the next elec on is due in Source: 2008 Survey on Monitoring the Paris Declara on Table 19: Budgeted and disbursement varia on, 2006/07 US$ millions Donors scheduled disbursements to the government sector Disbursements included in government budget Total disbursed aid Of which: for government sector Government accounted as received World Bank EC DfID GFATM AfDB Germany Sweden France Denmark Japan Others Total

40 A Globally Compe ve And Prosperous Kenya Box 4: Vision 2030, the medium term plan and spending needs FOUNDATIONS Physical Infrastructure: Increase the stock of physical infrastructure, so that there are 64,500km of well maintained roads. A total of Ksh 186 billion is an cipated for road construc on and upgrade during The PPP policy is expected to expedite private sector par cipa on; e.g. there will be concessions for main toll roads to be built by the private sector. There are plans for a new transport corridor to be built, linking Lamu, Ethiopia, Southern Sudan and Somalia, to a second port to be constructed at Lamu through Build, Own, Operate and Transfer (BoT) arrangements at a cost of US$ billion. A free port in Mombasa will also be developed. The other major transport infrastructure projects will be the development of a Rapid Bus and light rail system in the Nairobi Metropolitan area, which is expected to serve as a prototype for the other main urban areas in the country. Energy: Expand the network to connect a million households at a cost of Ksh 84 billion in five years; connect Kenya to the Southern Africa power pool through Tanzania at a cost of US$ 110 million in two years; and undertake other projects such as geothermal, solar and wind power. Provide solar generators to 74 public ins tu ons at Ksh 180 million. ECONOMIC PILLAR Tourism: Facilitate the crea on of 3 resort ci es. Agriculture: Irrigate an addi onal 1.2 million hectares of land for crop produc on, of which 600,000 in Arid and Semi-Arid Lands (ASALs), and establish five disease-free zones. Manufacturing: Set up a Special Economic Cluster ini ally in Mombasa, and a second in Kisumu for manufacturing establishments, and set up at least five Small and Medium Enterprises (SMEs) Industrial Parks and Specialized Economic Zones. Wholesale and Retail Trade: Build a free trade port at Mombasa, construct wholesale, retail and hawkers markets in selected urban areas. ICT and BPOs: Establish a Business Processing Outsource Park and digital villages; implement the Na onal Terrestrial Fibre Op c Network Project. Financial Services: No big cket expenditure items iden fied. SOCIAL PILLAR Educa on: Construct and fully equip 560 secondary schools, build at least one boarding primary school in each cons tuency of the ASAL districts, recruit an addi onal 28,000 school teachers. Health: Introduce community-level health units, supported by a well trained cadre of Community Owned Resource Persons and Community Health Extension Workers. Water: Rehabilitate and protect forests in five water towers, implement water storage and harvesting program, construct two large mul -purpose dams with capacity to store 2.4 billion cubic meters, and develop the sanita on and urban sewerage programs. Gender, Vulnerable Groups and Youth: Increase the Women Enterprise Fund and the Youth Enterprise Fund; facilitate the training of young people in technical, voca onal and entrepreneurial skills, expand polytechnics; and employ youth in labor intensive projects (e.g. road building and tree plan ng). Housing: Provide incen ves to the private sector to build 200,000 housing units annually by 2012, and to individuals through a secondary mortgage finance corpora on. Encourage local authori es to provide serviced land and construct low cost housing units through PPP arrangements. POLITICAL PILLAR Governance, Peace Building, and Conflict Management: Pilot a na onal CCT/Camera Surveillance project in Nairobi, Mombasa, Kisumu, and Nakuru; build an addi onal 20,000 housing units for Police Staff Housing; and establish a Na onal Security Data Center and roll out the Na onal Community Policing Ini a ve. An Independent Truth, Jus ce and Reconcilia on Commission (TJRC) and a permanent Ethnic and Race Rela ons Commission of Kenya will commence in 2008/09. 30

41 A Globally Compe ve And Prosperous Kenya There is a disconnect between the na onal plan and the sector plans, causing confusion in spending priori za on. The sectors and line ministries are supposed to adjust their plans, projects and programs to fit with Vision 2030, since it was not a bo om-up development and, therefore, there is a difference between sectoral plans and priori es and the na- onal plans. Given the approach taken, it would make sense for the clearly iden fiable priori es with public finance implica ons to be brought to bear on the budget process. Instead, the medium-term plan, the sectoral plans and the development budgets are not always consistent. For example, while the Vision 2030 and medium-term plan iden fied five key water catchment areas as priority areas, the dra Environment Sector Medium Term Strategy for and the budget itself give the five areas very low priority for public finance. Since the printed es mates do not show project level informa on, the classifica on system is not useful for iden fying which expenditures are aligned to which of these na onal (or sectoral) objec ves. There is no men on, for example, of Vision 2030 aspira ons for a new transport corridor linking Lamu to Ethiopia in the Ministry of Roads sec on of the printed budget es mates. Despite the infrastructure focus of the na onalgoals, the recent development budget does not show an increase in the share of resources to the sector. It is possible to determine the shares of the development budget according to the MTEF Sector Classifica on. There is li le change in overall sector composi on of spending in the last two budget cycles (see Figure 19 below). The need for significant new public investment in physical infrastructure, while increasing in real terms, is not evident in a sizeable increase in the share of total spending, which has actually fallen in the 2009/10 budget. This might reflect the fact that the budget process involves a fair amount of back and forth between the center and the line ministries, which makes the se ng of strategic direc ons and s cking to them much more difficult. It also reflects, to some degree, the extent to which the environment has changed since Vision 2030 was dra ed, and the need to focus on equity issues and on the challenges in the social sectors. Figure 19: Sectoral budgets 2008/09 and 2009/10 Source: Government of Kenya, Printed Es mates Key: P. Admin.-Public administra on; HRD- Human Resources Development; EWS- Environment, Water and Sanita on; RIT- Research, Innova on and Technology; GJLOS-Governance, Jus ce, Law and Order; TTI-Trade, Tourism and Industry. 31

42 A Globally Compe ve And Prosperous Kenya What is also missing from the budget process is the link to the financing gap required from the private sector through PPP arrangements. It is not clear if line ministries are required to priori ze the PPP projects or to explore this as a funding op on as part of the budget process. This is the subject of a parallel report being produced by the World Bank, and an area for future work since there is much more scope for proac vity that could unlock the unrealized poten al. The government, for example, could ensure that development expenditure is aligned with the private investments required to produce the flagship projects and other Vision 2030 aspira- ons, and that the scope for leveraging private finance is ac vely considered during the budget alloca on process. Project Management Poor project management is at the heart of the poor development budget execu on rates. World Bank (2007b) found that problems of large delayed payments, pending bills and arrears weakened local (technical) absorp ve capacity. Local contractors, par cularly in roads, suffer from poor cash flow and were unable to successfully complete projects on me, resul ng in reduced quality and scope, as well as me and cost overruns. Overall, projects financed 100 per cent by the government are roughly twice as likely (34 percent of projects) not to be completed on me as compared to the externally funded projects (16 percent of projects). Within infrastructure sectors, water resource management projects seem to face more difficulty for mely comple on (for both government and externally funded projects). The average delay in water resource project implementa on is three years.¹² It is, therefore, not surprising that there are such capacity constraints in light of the substan al increase in spending under the development budget observed in earlier sec ons and shown in Figure 14 in the execu ve summary. It appears that the weak absorp ve capacity will neutralize the poten al impact of increased spending. Delays in project implementa on, donor disbursements and u liza on of resources are common and arise from a large number of complex problems. Thirty eight weaknesses were iden fied by a joint Government of Kenya-World Bank review (2009) of the disbursement difficul es in IDA-financed projects. Not all of these were issues for all of the projects but in total illuminate a very challenging environment in improving the effec ve implementa on of exis ng projects (See Box 5). Project execu on takes a number of detailed steps, which require me. However, several specific problems have led to incredibly long delays in disbursements on both the government and on the development partner sides. The government s internal audit department and the World Bank point to the following overriding weaknesses: Long procedures and slow processing of documenta on. Accoun ng challenges as a result of: (i) incomplete financial management systems; (ii) inadequate accoun ng for pooled donor funds; and (iii) unsa sfactory value for money, fraud and poten al corrup on concerns in projects. Lack of project staff and high staff turnover on both sides. Complex project design, which hampers implementa on. Doub ul project sustainability. The vast majority of contracts in the roads por olio were awarded without the requisite depth of planning, feasibility study, site inves ga ons, proper design details, and engineering documenta on (see World Bank 2007c for a review of the project management issues in the roads sub-sector). The problems iden fied in this report include: rushed ini al designs; delays in star ng project implementa on because of poor ini al prepara on prior to budget entry, leading to me and cost overrun; insufficient field inves ga ons and oversight mechanisms resul ng in inappropriate engineering interven ons; over-designing; and waste of resources in some cases. The report also notes irregulari es such as the issuing of varia on orders to ini ate new projects. Typically, these varia ons would be added onto ongoing contracts, regardless of government policies, the project s own plans and appraisal, and the ca ¹² World Bank (2007b). 32

43 Box 5: Disbursement Problems: The Case of the World Bank A Globally Compe ve And Prosperous Kenya The problems causing disbursement delays are complex, with some ins tu on-specific, while others are project-specific. 38 dis nct problems, iden fied by a joint Government of Kenya-World Bank review (2007c) of IDA financed projects are summarized in six major groups below. Problem 1: Limited understanding of the procedures. The disbursements procedure for each project is defined and documented in a Disbursement Le er, which forms part of the legal loan agreement between the government and the Bank. This le er is normally a key part of project nego a ons. However, those nego a ng do not always make it available to those implemen ng the project. As a result, implementers of projects are not always aware of what needs to be done before a disbursement can be signed-off. Problem 2: Not having the right disbursement methods for the project. Several disbursement methods (e.g. direct payments, reimbursements, transfers to designated accounts, advances, etc) are available for projects, and they have differing purposes, implica ons, requirements and complica ons. Careful selec on during project design supports future success in implementa- on and disbursements. In par cular, not matching a project s specific environment, plans, and capacity to the most appropriate or suitable disbursement methods holds back project poten al. Simpler methods are o en not used; the more complicated procedures are more ac vely pursued. In contrast, the Kenya Educa on Sector designed a one-tranche disbursement for the first year of the project (advance payment followed by spending accoun ng), which has supported quick disbursements. Problem 3: World Bank s and the government s accoun ng requirements are not always compa ble. Frequently, Bank teams wish to see accounts that show how the finances are aligned with a project s specific objec ves and with the classifica on (e.g. goods, works and services) of the procurements made. However, this approach is not consistent with government s classifica on system. Moreover, it is difficult to dis nguish the Bank s funds from that of other donors in basket arrangements, or even from government s own funds. Bank-supported disaggregated policy objec ves are o en not the result alongside which goods and services are procured. Problem 4: Inability to suitably account for funds due to a weak repor ng capacity. Since disbursement approvals are stringently dependent on correct repor ng, some delays in disbursement are driven by weak repor ng capaci es. Cases of disbursement requests that have been rejected and amended over and over again before approval is finally possible, are frequent. Problem 5: Compounded delays in transferring funds to the project. Projects do not receive an- cipated funds on me if a delay occurs in any one, or more, of the following places during the protracted request-transfer process: (i) if Task Managers or the loans department holdup approvals at the World Bank; (ii) ) if the Central Bank (once funds are disbursed by the Bank) does not release funds from the interna onal reserve to the Treasury, and for the PIUs, fast enough; (iii) if the project operates an off-shore account, which takes much more me to bring the funds into Kenya; (iv) if Treasury does not release funds from its cash reserves to line ministries fast enough; and (iii) if line ministries decide to allocate in piecemeal amounts to project implementa on units. Kenyan parastatals, surprisingly, do not suffer as much from these delays. This is largely because they are much more independent of central government, and they hold special accounts that are directly managed by the Project Implementa on Units (PIUs). Problem 6: A Bank structure that s unhelpful for task teams. The Bank s internal organiza onal structure does not allow for daily interac ons between the financial management specialists, the procurement specialists and the loan officers (responsible for disbursements). In part, this is consciously designed to ensure a clear division of accountabili es between the approving, spending and release of funds roles. However, in prac ce, this generates considerable confusion and delay for the task teams. 33

44 A Globally Compe ve And Prosperous Kenya pacity of the contractor. The roads sector review also notes that there are no independent gate-keepers to screen project proposals, and decide whether a project meets the necessary level of prepara on or economic jus fica on. Furthermore, it is not easy to make judgments on the projects iden fied or selected when an overall road sector development policy and plan is not available. Implementa on of public investment projects are o en delayed by the lengthy and late started procurement processes. Since a project is not approved un l it is approved by Parliament, no substan al project prepara on, including procurement processes, can be effected in advance. Ministries, for example, cannot finance preparatory feasibility studies un l the budget is approved. This means that the procurement processes, which typically take several months from ini a on to conclusion, are launched during the project s core implementa on period, thus limi ng the project s spending ability from year one. Spending is also compromised when annual budget ceilings do not allow a project to undertake all its required procurement processes. And even if supplementary funds are requested (typically submi ed half-way through a fiscal year), the supplementary budget is frequently approved at the end of the year (usually during the last two months). This, of course, does not give the project me to ini ate, complete and reflect any procurement spending before year-end. Finally, where development partners are involved, addi- onal procedures and actors further complicate projects procurement implementa on and spending capacity. Inadequate capacity in project management and lack of technical staff are also key causes for poor implementa on of the development budget.¹³ The crea on of new administra ve districts has spread the limited resources available even thinner, par- cularly in provincial administra on and services in educa on, health and agriculture. There is a high turnover of staff and/or understaffing in many project units, and the few technical and project management staff available are stretched over all projects. Inadequate project appraisals or designs lead to poor implementa on and spending. Projects in the por olio that may not have been well appraised, or aligned with ministerial priori es show problema c implementa on and spending, largely due to: (i) insufficient a en on given the project s low policy/ priority status; or (ii) the difficul es in implementing a poorly appraised and designed project. Other projects that are equally problema c are those that spread their budgets too thinly across geographical loca ons to accommodate the social-poli cal concerns of poten al beneficiaries. Improving the Role of the Development Budget A piece meal or quick-fix approach will not resolve the development issues iden fied above. Instead, a comprehensive strategy is needed and one that addresses, among other considera ons, the following specific recommenda ons and conclusions from this report: (a) Ensuring consistency between sectoral and na onal priori es. There is s ll a need to review the consistency between the medium term plan and the sectoral plans. In order to get spending in line with priori es, it is first important that the priori es are clear, and currently they are not. (b) Strengthening spending alignment with na- onal priori es. The various departments within Treasury and in the Ministry of Planning and Na onal Development departments need to work together to ensure that budget alloca ons are consistent with the stated na onal priori es. For example, the Vision 2030 flagship projects that are brought into the five-year medium term plan should be screened and priori zed for new discre onary development budget alloca ons. (c) Improving the role of public finance as leverage. The extent to which new projects receive public financial backing needs to be looked at in the context of leveraging private financing. The PPP policy needs to be finalized, and the guidance for taking this forward brought into the annual budget cycle. (d) Strengthening spending alignment with sectoral priori es. Within sectors, the SWGs are required to make the necessary trade-offs and provide advice on sectoral policies. Treasury should discon nue the prac ce of reoreopen- ¹³ Kirira (2009) 34

45 A Globally Compe ve And Prosperous Kenya ing up budget alloca on discussions with individual line ministries once the SWGs have completed their recommenda ons. Instead, any addi onal resources should be given to the SWGs to decide on priority realloca ons. (e) Proac vely dealing with stalled projects. The progress with implemen ng stalled projects should be maintained, and the remaining ones reviewed with a view to hal ng projects that are no longer a priority. (f) Improving implementa on of sectoral strategies. The SWGs should be required to report to the Economic and Budget Steering Commi ee (EBSS) on the extent to which their budget submissions meet the targets laid out in their own sector plans, and the na onal medium-term plan. (g) Introducing more effec ve project screening and appraisal. A standard format and screening process for the proposed new projects is needed in the budget circular. This should be designed in a way that ensures that all proposed new projects are fully assessed for their coherence with sectoral plans, the MPER recommenda ons and the iden fied priori es across the sector (rather than across ministries) before inclusion in the budget. (h) Improving and standardizing project appraisal and design. A robust economic appraisal, and one that draws relevant lessons and applica ons from other parts of the country and the world should be a minimum requirement for all proposed development projects. This, and other cri cal requirements, could be presented in clear and mandatory guidelines that are issued alongside the budget circular. Examples of other appraisal requirements that would be listed in the guidelines include: informa on on es mated project costs (both fixed and recurrent over a reasonable meframe), the likely benefits of the project, the alterna ve interven ons considered, the environmental and social impacts (if any), and the staffing and implementa on arrangements. Such guidelines, including templates for some of the informa on sought, could be adapted from countries that have already adopted the approach. (i) Improving the pace of implementa on. Project budgets, and especially those with mul -year implementa on and expenditures, should be protected and approved. There should be an assump on in the MTEF, going forward, that exis ng commitments will be honored and this should come off the top before ceilings are issued for the development budget. This would help get rid of projects that have stalled or are taking too long to deliver because of insufficient budget alloca ons. (j) Accelera ng implementa on and execu on. Prepara on of technical drawings, tender documents and feasibility studies should all start long before budget approval so that implementa on can begin at the start of the fiscal year, and at the start of the project. 35

46 3. PRO-POOR SPENDING IN KENYA: A REVIEW OF THE TARGETED SUBSIDY PROGRAMS Background Recovery from post-elec on violence experienced in the first quarter of 2008 was muted by new exogenous shocks later in the year, including the drought and global financial crisis. Economic growth slowed to 1.7 per cent in 2008, down from 7 per cent in 2007, and is projected to be only 3 per cent in Infla on rose ini ally due to the violence-driven disrup ons to supply chains and rising global energy prices, and more recently due to rising interna onal food and fer lizer prices. Overall month on month infla on was 16 per cent in 2008 and 9.4 per cent in 2009.¹⁴ Kenya remains vulnerable to further domes c and external shocks, as it does not have assured sources of external finance from either official or private sources. Food supply shortages will maintain infla onary pressure in 2009, at least un l the first harvest, which takes place in the last quarter of the year. Increased imports of maize will deplete foreign exchange reserves at a me when export earnings, remi ances and capital inflows are all dampened by the nega ve impact of the global financial crisis. As a result of the combined effect of high food prices, low economic growth, and unemployment, the number of poor and vulnerable Kenyans will increase un l the economy is on a path to full recovery. In 2008, for instance, real GDP growth of 1.7 per cent and popula on growth of 2.7 per cent meant a decline in per capita incomes of around 1 per cent.¹⁵ In 2006, poverty es mates based on the Kenya Integrated Household Budget Survey (KIHBS, 2006/05) indicated that 16.3 million Kenyans were food-poor and could not meet the cost of a basic food bundle. The World Bank s Kenya Poverty and Inequality Assessment (KPIA) showed that although poverty declined between 2000 and 2006, the poverty incidence is s ll high at 47 per cent. Furthermore, rural poverty is much higher at 49.7 per cent compared to 34.4 per cent in urban areas. The na onal numbers mask important regional differences; in the Coast and North Eastern provinces, poverty incidence is es mated at 70 per cent and 74 per cent, respec- vely, compared to 22 per cent in Nairobi and 31 per cent in Central Province.¹⁶ However, although North Eastern Province has the highest level of poverty, the province only contributes 4.9 per cent to ¹⁵ The popula on growth rate in 2008 is es mated at 2.7 per cent (Economic Survey, 2009). ¹⁶ Current es mates by the Kenya Food Security Group (KFSG) indicate that about 10 million people in Kenya are food insecure and vulnerable to domes c and external shocks. The es mates indicate that out of the 10 million, 4.1 million are in urban areas. These es mates are compiled based on the Early Warning System (EWS), which is based on rainfall pa erns, and updated every six months. ¹⁷ Defined as consump on levels below the Minimum Dietary Energy requirements (MDER), i.e. the amount that is considered adequate to meet individual daily energy needs for light ac vity and good health for age and sex. ¹⁸ Food Insecurity Assessment in Kenya: Based on Kenya Integrated Household Budget Survey 2005/06, KNBS