1

Performance-Oriented Budgeting in Korea:

Evidence and Lessons

June 2008

Nowook Park∗∗ Korea Institute of Public Finance

∗∗ Center for Performance Evaluation and Management, Korea Institute of Public Finance, 79-6 Garak-dong Songpa-gu, Seoul 138-774, Republic of Korea. Email: npark@kipf.re.kr

2

1. Introduction

Recently, efforts to enhance links between program performance and budgeting have been in full swing not only in many OECD countries but also in a significant number of developing countries. A survey of OECD countries reports that 70% of countries include non-financial information in their budget documents. Emerging economies, such as China and Brazil, are also in the process of introducing performance information into their budget process. Performance-oriented budgeting is not a new initiative and has a long history in advanced countries. The first serious attempt to introduce performance budgeting occurred in the United States during the 1960s in a program called Planning Program Budgeting System (PPBS). Since then, various initiatives have tried but they were short-lived along with the administrations that initiated them. However, performance budgeting has not died out and has been gaining momentum in many countries since the 1980s. While many OECD countries engage in new initiatives on performance-based budgeting, Korea also exhibits very strong movement towards performance-based budgeting.

The Korean case can be considered as a prominent example of recent efforts to introduce performance-based budgeting into the government. Since the Korean government started to introduce and quickly implemented performance management system as one component of a comprehensive fiscal reform package, the Korea case offers us a very interesting opportunity to observe and learn lessons. An advantage to these multi-faceted efforts is, if coordinated properly, it creates a favorable environment for an effective performance management. The downside to these efforts is that it demands huge resources that may be readily available to many countries.

During 2003-2007, the Korean government implemented major reforms in its budget process. They are called as “Four Major Reforms in Public Finance”: Medium-term fiscal plan (national fiscal management plan), top-down budgeting, performance-oriented budgeting and digital accounting system. These reforms efforts are introduced as a response to the deteriorating fiscal situations of the Korean government. After the Asian financial crisis in the late 1990s, public debt was dramatically increased although it is not quite high compared to the average of OECD countries. Also the acceleration of economic inequality called for the provision of social safety net for economically disadvantaged people. As population aging is progressing at an unprecedented speed, a burden of supporting older people will become harder to bear as time goes on. These concerns motivated the Korean government to take on reforms in public expenditure management.

3

The medium-term fiscal framework as develop to give a 5-year picture of government spending based on the projected economic growth. It is supposed to provide a big picture to direct the allocation of funding among the major 14 sectors. Based on this projection, funds are allocated to each line ministry by a top-down approach. Each line ministry receives funds by a fixed ceiling and has autonomy to decide detailed spending plan with the ceiling. In return for the autonomy enjoyed by each spending ministry, a performance management system is introduced to examine the performance of spending programs and enhance links between budgeting and performance. To provide infrastructure for eh aforementioned three-pronged reform programs, digital accounting system, including program budgeting and accrual accounting, was developed in 2005 and will be in full implementation by 2009.

This paper tries to describe the current initiative, to examine its effect, and to analyze issues involved in the implementation of the performance management system of budgetary programs in Korea. This paper is organized in the following. First, it describes the current system of performance-oriented budgeting in the Korean central government. Second, it examines its results and effects on the budget allocation and the management of spending programs. Third, it discusses challenges of implementing performance-oriented budgeting and future directions.

2. Description of Performance –Oriented Budgeting in Korea

This section describes the history of performance management of budgetary

programs in Korea.1 The first attempt to manage budgetary programs occurred during 1999-2002. In 1999, research on the existing various performance management systems around world was done to find a viable way to introduce it. Based on the research, a pilot project started on the selected divisions and branches of 16 ministries in 2000 and expanded to 22 ministries in 2002. It was designed after the GPRA (Government Performance and Results Act) in the United States and named as "Performance-based Budgeting." Under this project, the spending ministries produce annual performance plans and reports. The response to the pilot project was lukewarm among the ministries and it ended with the incumbent Administration.

By drawing upon the previous experience, "Performance Management System of Budgetary Program (PMSBP)" was introduced to the 16 leading ministries in 2003. Its

1 This section builds upon Park(2006a).

4

implementation was not limited to selected divisions or branches within each ministry but applied to the entire ministry. It started with developing a mission statement, strategic goals, performance goals, and performance indicators in the annual performance plan. In 2003, performance information on 30% of budgetary programs has developed for the 16 leading ministries. In 2004, another 16 ministries developed performance information on 30% of their budgetary programs and the 16 leading ministries developed performance information on 100% of their budgetary programs. In 2005, 22 ministries submitted an annual performance plan. However, expansion of PMSBP to other ministries is withheld because the Office of Government Policy Coordination (OCP) will take over PMSBP by expanding it to the entire ministries/agencies including their entire programs.

PMSBP was designed after the GPRA in the United States which aims to introduce performance-based budgeting. What distinguishes the GPRA from previous attempts is the fact that it is initiated not by the Administration but by the Legislature. The GPRA was enacted in 1993 and implemented from 1997, and it requires each agency to submit its 5-year strategic plan, updated in every 3rd year, an annual performance plan and an annual performance report.

Differences between PMSBP and GPRA were the following: (1) strategic planning was not introduced in PMSBP. (2) Every program was not included in PMSBP while GPRA covered every program including personnel management and operation programs. (3) PMSBP developed performance information by employing a bottom-up approach, while GPRA utilized a top-down approach.

These differences are a reflection of the Korean context where performance-based budgeting was being implemented. Programs mainly related to personnel management and routine operation were not included. That is because there was another performance management system handled by the Office for Government Policy Coordination (OPC) and the Ministry of Planning and Budget (MPB) needed to avoid the perception of duplication. MPB took an easy way of benchmarking GPRA by skipping strategic planning and employing a bottom-up approach in that each line ministry does not have to consider reorganization of existing programs. Ministries/agencies developed performance goal and strategic goals from existing programs. As a result, one of the important components of performance management was lost. However, in 2006, the OPC introduced "Integrated Evaluation System of Government Affairs (IESGA)" to coordinate existing evaluation systems conducted by various ministries, such as, MPB, Ministry of Government Administration and Home Affairs, OPC, and Civil Service Commission, and strategic planning was

5

introduced as a component of requirements, even though true sense of strategic planning has rarely been realized among line ministries.

In 2005, MPB introduced "Self-Assessment of Budgetary Programs (SABP, sometimes called as Korean PART)" to enhance links between performance evaluation and budget allocation. It is designed after Program Assessment Rating Tools (PART) in the United States. While PMSBP relies on the small number of key performance measures for the performance management, SABP collects more detailed information on budgetary programs. SABP examines planning, management and results of budgetary programs by a checklist designed by MPB. Line ministries/agencies evaluate their programs by answering the questions provided by the checklist. It is designed to evaluate all the budgetary programs in 3 years by evaluating about 30% of programs every year. More than 1,700 programs have been evaluated between 2005 and 2007 and their results are reported in section 3.

In the summer of 2005, MPB reorganized its departments and created a bureau of performance management to give institutional backbone to the performance management system. Within the bureau, the team for the performance management system is responsible for the PMSBP2 and SABP while program evaluation team handles "in-depth program evaluation" which was introduced in 2006.

In addition to PMSBP and SABP, MPB introduced "in-depth program evaluation" in 2006. MPB selects programs to be evaluated based on the SABP results and inputs from experts, interest groups, and agencies. The selected programs will be subject to in-depth evaluation in order to make decision based on more conclusive evidence. While the experience of program evaluation is widely available in many OECD countries, Korea has little experience in the field. It remains to be seen how effectively program evaluation will be conducted.

Since Korean government introduced in-depth program evaluation in 2006, it is equipped with a complete set of performance management system of budgetary programs. [Figure 2-1] shows the framework of performance management system of budgetary programs in Korea. PMSBP will play a role of monitoring by examining performance indicators, SABP will provide more detailed information on the performance of budgetary programs by checklist, and in-depth program evaluation will offer more conclusive performance information on selected programs. Regarding the unit of analysis, PMSBP focus on the level of programs and sub-programs, SABP

2 PMSBP renamed as Performance Goal Management System of Budgetary Program (PGMSBP) in 2006.

6

evaluate at the level of sub-programs.3 The unit of analysis for in-depth program evaluation varies with evaluation issue of a specific program.

In December of 2006, “National Financial Management Law” was enacted where “Four Major Reforms in Public Finance” was stipulated, including performance management system of budgetary programs. Inclusion of performance-oriented budgeting into the law is expected to give a stable environment for developing effective performance management system, because it takes time to produce relevant performance information and to change spending ministry’s practice managing budgetary programs.

[Figure 2-1] Framework of Performance Management of Budgetary Programs * Management of Performance

Goals and Measures (Strategic Goals-Performance Goals-Program)

- Monitoring (Annual Performance Plan

and Report))

⇓

* Self-Assessment of Budgetary Programs (Over 1,700 Programs)

- Review (Checklist)

⇓

* In-depth Program Evaluation (In-depth evaluation of selected programs)

- Evaluation (In-depth Evaluation)

3 The Korean government has 1,043 programs and 3,594 sub-programs in 2007.

7

3. Results of Performance-Oriented Budgeting during 2005-2007

Even though, as discussed in the previous section, the Korean government has three components of performance management system of budgetary programs, only SABP (a.k.a. KPART) has been actively implemented and has enough data to report its results. Thus, in this section, the results and effects of the SABP during 2005-2007 will be mainly discussed. Each year 555, 577 and 584 programs were evaluated by going through 3 rounds of communications between ministries/agencies and MPB. They assessed the program’s performance of fiscal year 2004, 2005 and 2006. Statistical Summary of evaluation results and utilization of evaluation results in budget process will be discussed.

3. 1. Statistical Summary of SABP Results Distribution of effectiveness rating given to each program is presented in <Table 3-

1> and [Figure 3-1]. The proportion of programs rated as ineffective is 15.7% and a majority of the programs are rated as adequate in 2005. 5% of programs are evaluated as effective.4 The distribution of program rating did not change much between 2005 and 2006, but it was greatly improved in 2007: the ratio of ineffective programs is dropped to 5% and that of effective programs are doubled. The significant improvement in programs’ ratings in 2007 is attributable to the improvement of program’s performance and agency’s learning effects from a couple of years’ experience with SABP.5

SABP consists of 3 sections which are program planning, management, and results. Program management section can be divided into two parts; (1) program’s rationale and design (2) performance plan. Among 15 common questions in the checklist, 7 questions are about program planning, 4 questions are about program management, and the remaining 4 questions are about program results. [Figure 3-2]-[Figure 3-4] report the distributions of each section's score. The distribution of scores in the planning section is skewed towards the upper bound. Particularly that in the

4 Compared to the experience of the United States, it is very surprising, because 50% of programs were given "result not demonstrated (RND)" rating at the first-year PART rating. If you listen to the inside story, however, a different picture emerges. According to the insiders, initially 70% of programs were judged as "result not demonstrated (RND)" due to lack of performance information. For some reason, MPB decided not to create RND rating and gave more generous evaluation and came up with the ratings in <Table 3-1>. 5 This could be an effect from change of standard in evaluation, even though there is no convincing evidence.

8

rationale and design section shows not much variation among evaluated programs and is close to perfect score.6 In contrast, that in the performance plan section shows great variance among programs, which suggests that developing relevant performance is not easy. Its distribution of the management section shows that many programs obtain perfect scores. Its distribution of the results section exhibits skewed distribution towards the lower bound.

<Table 3-1> Results of SABP in 2005

(Number of Programs, %)

Total Effective

(>85)

Moderately Effective

(84~70)

Adequate

(69~50)

Ineffective

(<50)

2005 555 (100) 28 (5.0) 100 (18.0) 340 (61.3) 87 (15.7)

2006 577 (100) 30 (5.2) 94 (16.3) 388 (67.2) 65 (11.3)

2007 584 (100) 66 (11.3) 139 (23.8) 348 (59.6) 31 (5.3)

[Figure 3-1] Distribution of Total Score

0

50

100

150

200

250

300

0 10 20 30 40 50 60 70 80 90 100 Totalscore

Frequency

2005 2006 2007

6 It may be a reflection of the fact that SABP is an evaluation within the Administration and, as a result, it is hard to question existing programs’ rationale itself.

9

[Figure 3-2] Distribution of Scores in Planning Section

Distribution of Scores in Rationale and Design Section

0

100

200

300

400

500

600

0 3 6 9 12 15 Designscore

Frequency

2005 2006 2007

Distribution of Scores in Performance Plan Section

0

50

100

150

200

250

300

0 3 6 9 12 15

Performance planscore

Frequency

2005 2006 2007

10

[Figure 3-3] Distribution of Scores in Management Section

0

50

100

150

200

250

300

350

400

450

500

0 3 6 9 12 15Management

score

Frequency

2005 2006 2007

[Figure 3-4] Distribution of Scores in Results Section

0

20

40

60

80

100

120

140

160

180

0 5 10 15 20 25 30 35 40 45 50 Resultscore

Frequency

2005 2006 2007

11

Descriptive statistics by each section are summarized in <Table 3-2>. The performance plan section in the planning section and results section exhibits relatively low mean scores compared to that of the program design and management section. Particularly programs rated as ineffective received relatively low scores in the performance plan and results section. Considering the fact that the scores in the results section are linked to that in the performance plan, lack of an adequate performance plan may be the most deciding factor in determining a program's rating.

12

<Table 3-2> SABP Results by Section SABP Results by Section

Planning (30)

Year Total

Score Rating

# of

Program Planning

(30)

Program

Design (15)

Performance

Plan (15)

Management

(20)

Results

(50)

Total 555 23.05 13.75 9.30 15.13 21.94

>85 Effective 29 29.03 14.03 15.00 17.14 42.42

70~84 Mod.Eff. 104 26.37 13.87 12.50 16.59 32.31

50~69 Adequate 337 22.50 13.94 8.56 15.30 19.74

2005

50> Ineffective 85 19.12 12.77 6.35 12.00 11.02

Total 577 22.94 14.27 8.68 14.72 22.30

>85 Effective 30 29.63 14.80 14.83 16.72 42.41

70~84 Mod.Eff. 96 27.72 14.49 13.23 15.37 33.17

50~69 Adequate 387 22.01 14.46 7.56 15.12 19.82

2006

50> Ineffective 64 18.15 12.52 5.63 10.36 11.13

Total 584 23.40 14.20 9.20 15.50 27.10

>85 Effective 69 29.66 14.88 14.78 16.74 43.10

70~84 Mod.Eff. 144 26.80 14.71 12.08 16.44 32.32

50~69 Adequate 340 21.30 14.15 7.15 15.27 22.71

2007

50> Ineffective 31 16.66 10.85 5.81 10.89 15.40

Note: Numbers in the parenthesis represent possible maximum scores.

<Table 3-3> reports detailed results on the crucial questions in determining program's

rating. Two important questions on the performance plan are about whether a program

has relevant performance indicators and reasonable performance targets. 39.8 % of

programs are evaluated as having no relevant performance indicators in 2005, and it did

gets worse in 2006 and 2007, where the ratios is dropped to 47.5% and 43.9%,

respectively. Regarding performance targets, this number goes up to 71.5% in 2005

and stays about the same level afterwards. It shows many programs do not have

adequate performance information yet.

Let’s examine what is the main reason for the improvement in program rating in 2007.

Since there is no noticeable change in the section of program design, performance plan

13

and management, improvement in the result section is the main source of improvement.

In <Table 3-3> significant improvement comes from questions on meeting the

performance targets (question 3-2), achieving citizens’ satisfaction (question 3-3) and

utilizing evaluation results in decision making process (question 3-4).

<Table 3-3 > Evaluation Results on Performance-related Questions

(Unit: #, %) Evaluation Results on Performance-related Questions

(Unit : %)

Year # of

Programs Question "Yes" "No" "Somewhat"

"Very

much"

1-6 Performance Indicator 60.2 39.8

1-7 Performance Target 28.3 71.7

3-1 Program Evaluation 62.7 37.3

3-2 Performance Results 9.5 10.5 62.2 17.8

3-3 Customers' Satisfaction 47.8 52.2

2005 555

3-4 Feedback 57.5 42.5

1-6 Performance Indicator 52.5 47.5

1-7 Performance Target 22.4 77.6

3-1 Program Evaluation 49.2 50.8

3-2 Performance Results 12.5 4.5 72.3 10.7

3-3 Customers' Satisfaction 57.5 42.5

2006 577

3-4 Feedback 55.8 44.2

1-6 Performance Indicator 56.1 43.9

1-7 Performance Target 28.7 71.3

3-1 Program Evaluation 47.6 52.4

3-2 Performance Results 14.4 1.4 62.1 22.2

3-3 Customers' Satisfaction 81.3 18.7

2007 584

3-4 Feedback 79.7 20.3

Source: MPB

14

3. 2. Utilization of Performance Information in Budget Process7

In this section, let’s examine whether evaluation results are utilized in each stage of budgetary process, which may be the ultimate purpose of the performance-based budgeting. <Table 4-1> shows links between evaluation results and budget allocation. First, let's examine whether agencies utilized evaluation results in their budget requests. In terms of absolute amount, programs rated as ineffective suffered loss of 16 billion won in total, compared to 2005 budget. Since this big loss may be affected by huge losses in several big programs, it would be better to examine average ratios of budget change to draw a more precise picture. For programs rated as ineffective, agencies requested, on average, 15% less than 2005 budget. It clearly shows that agencies consider evaluation results in their budget requests. In 2007 budget request, agencies also requested, on average, 15% less for ineffective programs compared to 2006 budget. However, in 2008 budget request, ineffective programs received minimal budget cut which is 0.1% budget reduction compared to the 2007 budget. It suggests the extent to which evaluation results affects agencies’ budget requests becomes smaller.

Second, let's examine whether MPB utilized performance information in their budget proposal. <Table 4-1> shows clear links between evaluation results and MPB’s budget proposal. Programs rated as ineffective suffered a loss of 12.7 billion won in total amount and a 25% loss on average in 2006 budget. While ineffective programs lost 24% on average in 2007 budget, they suffer 15% budget cut in 2008 budget.

In fact, MPB encouraged ministries/agencies to make use of evaluation results in their budget requests and announced that a 10% budget cut will be imposed on the programs rated as ineffective. The analysis on the data strongly suggests that agencies and MPB made a clear link between evaluation results and budget allocation. While the extent to which evaluation results are utilized becomes weak in 2008 budget formulation, they are still utilized by MPB.

Third, let’s examine whether the National Assembly utilized performance information. In contrast to the case of the Unites States, the Korean case shows that ineffective programs went through budget cut in the Legislature in both 2006 and 2007’s budget approval. In sum, performance information is reflected in every stage of budget formulation between 2005 and 2007.8

7 The following results are based on the methodology employed in Park (2008). 8 Since budget allocation is affected not only by performance information but also by other factors, such

15

<Table 4-1> Links between Evaluation Results and Budget Allocation (Unit : 100,000 won)

Links between Evaluation Results and Budget Allocation ('05 KPART) (Unit : 100,000 won)

'06 Difference Ratio

'05

Budget

(A)

Agency

(B)

MPB

(C)

Final

(D) B-A C-A D-A Ratio_1 Ratio_2 Ratio_3

Effective 15,600 24,948 18,955 22,489 9,348 3,355 6,889 0.52 0.38 0.50

Mod.Eff. 92,994 104,335 107,055 105,762 11,342 14,062 12,768 0.32 0.33 0.28

Adequate 208,066 204,473 195,625 201,214 -3,593 -12,441 -6,852 0.10 0.05 0.05

Ineffective 33,081 28,644 29,007 28,505 -4,437 -4,074 -4,576 -0.15 -0.25 -0.19

Total 349,740 362,400 350,642 357,970

Links between Evaluation Results and Budget Allocation ('06 KPART) (Unit : 100,000 won)

'07 Difference Ratio

'06

Budget

(A)

Agency

(B)

MPB

(C)

Final

(D) B-A C-A D-A Ratio_1 Ratio_2 Ratio_3

Effective 8,891 9,467 9,337 8,872 575 446 -19 0.11 0.10 0.06

Mod. Eff. 33,156 35,701 35,364 35,654 2,545 2,208 2,498 0.12 0.08 0.09

Adequate 297,180 296,769 290,481 289,969 -411 -6,699 -7,211 0.10 0.04 0.03

Ineffective 11,431 6,039 5,400 5,380 -5,392 -6,031 -6,051 -0.15 -0.24 -0.25

Total 350,658 347,975 340,582 339,875

Links between Evaluation Results and Budget Allocation ('07 KPART) (Unit : 100,000 won)

'08 Difference Ratio

'07

Budget (A) Agency (B) MPB (C) B-A C-A Ratio_1 Ratio_2

Effective 17,112 18,211 17,503 1,099 391 0.18 0.12

Mod. Eff. 266,051 295,121 291,319 29,070 25,268 0.20 0.13

Adequate 146,034 153,026 142,044 6,992 -3,990 0.28 0.15

Ineffective 3,870 3,457 3,066 -414 -805 -0.001 -0.15

Total 433,067 469,815 453,932

as program type, ministries’ characteristics, and demand for program, it is necessary to control these factors. Park (2008) found strong correlation between budget allocation and evaluation results after controlling other factors at every stage of budget process.

16

3. 3. Utilization of Performance Information in Line Ministries

So far evidence suggests that evaluation results are utilized in budget negotiation

process and reflected in final budget allocation. This may be stark contrast with other countries. In this part, it will be illustrated how performance information is utilized at line ministries and how line ministries are changing their management practices of service delivery. According to the survey conducted by the National Assembly Budget Office (NABO) in Korea, 70% of budgeting personnel in line ministries responded with neutral or positive answers to the question, “Are evaluation results from SABP useful in the preparation of budget request?” Considering tendency of negative responses from those who are subject to evaluation to the evaluation process, line ministries show quite high satisfaction level with usefulness of evaluation results. Many budgeting personnel in line ministries say availability of evaluation results made their job easier, because performance information facilitates communication between them and program managers.

Not only line ministries utilize evaluation results from SABP in their budget formulation, but also change their program management to enhance the effectiveness of service delivery. An example from the Ministry of Health, Welfare, and Family is one of the cases in point. The ministry has been running a program designed to provide various help to the unemployed low income family. The program started in 2000 and those who become economically independent after receiving the program’s assistance are less than 3%. Due to its poor performance, the program has been under great pressure particularly after the introduction of performance-oriented budgeting. In response, the ministry is in the process of reforming the program.

The program has been managed in the following way: fund ranging between US $120, 000 - $190,000 was allocated to 242 regional centers. There is only one center in each locality and amount of funds is decided by the number of supporting family. Thus there is no competition among the centers and flexibility of fund allocation is very limited. Recently, the ministry decided to run a pilot program based on performance-based management. There will be separation between buyers and suppliers of the program: the service provider will be selected based on the applicants’ proposal. Applicants’ eligibility will be extended to private firms as well as non-profit organization. The contract will be based on the performance of service providers. There will be fixed payment proportional to the number of family they serve. Then, two additional incentives will be provided: (1) if a serviced individual finds a job

17

meeting a certain standard, bonus payment will be given to the service provider, (2) if a serviced individual become independent of government subsidy for low income family, additional bonus will be provided to the service provider. To avoid possibility of cream skimming, differentiated amount of fixed payment will be allocated according to the characteristics of family. This drastic reform initiative aims to improve the effectiveness of service delivery and is resulted from increasing pressure from performance-oriented budgeting process. The pilot program will be run this year and, in two years, its performance will be evaluated to see whether nationwide implementation is warranted. Performance-oriented budgeting in Korea is still at the early stage, but it is gaining momentum and changing the way of program management in line ministries, as illustrated in the case of the Ministry of Health, Welfare, and Family.

4. Lessons9

This paper explained institutional background and a current system of performance-oriented budgeting in Korea. This paper also tried to show whether performance information is utilized in the budget process and how it is utilized in line ministry.

This paper concludes with lessons and suggestions for the performance-oriented budgeting in Korea. At present, Korea is in the early stages of implementing performance-based budgeting; it is therefore too early to form a definite assessment. However, it should be noted that having performance-based budgeting introduced as one component within a broader range of comprehensive reforms has helped to lower resistance and resolve institutional problems. At the same time, concerns have been raised that the concurrent implementation of multiple major reforms in itself imposes an inappropriately heavy burden on the government. For example, program budgeting and performance-oriented budgeting have been implemented by different units in the central budget authority and, as a result, some inconsistencies are found between program structure and performance plan.

Overall, performance information has had a positive impact on the budgeting process. As regards the impact of performance information on ministry/agency performance, it is too early to tell, but as illustrated in this paper, ministry/agency is changing the way of managing programs to enhance its service delivery.

9 It is mainly drawn from Kim and Park (2007) with modification reflecting current progress in Korea.

18

Despite Korea’s meager experience time-wise in introducing performance budgeting, some general lessons can be drawn. Reorganizing ministries/agencies and the budget structure needs to be done before introducing the performance system. In Korea, performance information is focused on individual program/projects, but cost information is not readily available because organizational units, program, and the budget structure need to be realigned so that they are consistent with each other. As a result, it has been extremely difficult to develop meaningful outcome measures and efficiency/effectiveness measures.

Korea’s short experience has also shown that in introducing the performance system, decision makers should be patient about reaping any benefits. There is a concern that decision makers in Korea may be more interested in introducing the performance system than in monitoring or improving it. If a particular country is accustomed to getting quick returns from the reforms, it will not be easy to develop and improve the system.

The political and administrative culture of Korea poses some idiosyncratic challenges that other countries hopefully need not consider. Regular rotation of assignments in the civil service may work against the capacity development of ministries/agencies. Lack of patience on the part of decision makers may force the MPB to take ill-advised, excessive measures in order to show quick results. It was partly because of this kind of pressure that the MPB felt forced to quickly implement a 10% budget cut for ineffective program.

Korea’s experience confirms that a performance system evolves over time and raises different challenges at each stage. At the initial stage, merely developing relevant information is the main challenge. As the performance system evolves, other changes become more important, namely behavioral change, such as how to get various actors to use performance information in the decision-making process, and how to monitor the performance of the performance system itself.

The lessons from the Korean experience can be summed up as follows. Make sure the infrastructure is ready for the reforms. Proper cost accounting and a solid program budget structure will greatly help to maximize the benefits of the performance system. There should be proper understanding of performance-based budgeting among the civil service. Otherwise, wasteful and distorting behavior may proliferate.

Looking forward, the major ongoing problem for Korea is the quality of performance information. More training and research is needed, along with a greater commitment to invest in collecting and organizing the information. Specifically, the analytical and administrative capacities of the MPB and ministries/agencies need to

19

improve. This may require reinforcement of units specializing in evaluation in both the MPB and ministries/agencies.

20

References Kim, John and Nowook Park (2007) “Performance Budgeting in Korea,” OECD

Journal on Budgeting, Vol. 7, No. 4, Paris. Park, Nowook (2008) “Does More Information Improve Budget Allocation? Evidence

and Lessons from Performance-Oriented Budgeting in Korea,” Working Paper to be presented at the Congress of International Institute of Public Finance, August, Maastricht, Netherlands.

Park, Nowook (2006) “Performance Management in the Korean Public Sector: Current

Issues and Future Directions,” Working Paper presented at the OECD Senior Budget Officials Meeting, April, Paris, France.