Performance Budgeting Equals Soviet-style Central Planning?

Posted by Eivind Tandberg.

A common criticism of performance and program budgeting efforts is that they lead to budget planning systems that are mechanistic, overly complex, and overloaded with useless data; and that performance data have little or no impact on actual decisions. Many critics compare performance-oriented budget mechanisms to the planning models developed by Soviet central planners, and claim they are equally useful (or useless). Is there any truth to this claim? This blog provides a brief discussion of the claim in relation to three specific dimensions of performance budgeting: the historical roots, the concepts and the practical implementation of these concepts.

Historical Roots

 

Performance budgeting, under the original designation of program budgeting, was pioneered in the United States in the 1950s, and applied quite widely in OECD countries in the 1960s and 1970s. This form of program budgeting is clearly inspired by centralized planning. This was a period when governments believed that they could plan and manage the development of societies in great detail, and that it was important that they do so.

In Europe in particular, Soviet 5-year plans were a major inspiration for the new planning models. At this time, the underlying flaws in the central planning model were not yet generally apparent, and a number of European adopted elaborate government planning mechanisms, often based on macroeconomic models. Economists, engineers and social scientists provided the necessary input, and the ambition was to use dynamic programming to define socially optimal development paths. In order to ensure that government services contributed to the optimal development path, it was necessary to define programs and outputs and to assign (or compute) shadow prices to these outputs. According to some of the contributors, the main challenge for thecreation of an “optimal society” would be the availability of sufficient computing power and correct shadow prices.

Not surprisingly, 1960s' program budgeting attempts were not particularly successful, partly for the same reason that the Soviet planning model proved to be inadequate. Some countries produced elaborate sets of programs, documents and indicators, but the process was often mechanistic and decoupled from actual decision-making. Even more importantly, managers had limited opportunities and incentives to implement the programs in a way that produced the programmed results. In model language: the systems were often over-determined. Civil service management was hampered by a number of rigidities in deploying resources and managing personnel; tracking of indicators and results was not well developed; and there were few explicit rewards or sanctions related to results. Managers often perceived the program budget efforts as a passing fad, and paid little attention to it.

 

Modern Performance Budgeting 

There was limited progress in performance budgeting for several years. However, during the 1980s some countries began introducing approaches to performance budgeting that addressed the shortfalls of the initial program budgeting efforts. This was part of a broader renewal of public sector management aimed at improving performance and accountability, by reducing rigidities and micromanagement and providing much stronger incentives for managers and staff to improve performance.

This new approach to performance budgeting was clearly different from Soviet-style central planning (and from the initial forms of program budgeting). Table 1 provides a cursory, simplified comparison of some key elements of the two approaches.

 

Table 1: Comparison of Concepts

	Soviet central planning	Performance budgeting
Time frame	5 years	3 – 4 years, rolling
Coverage	Whole economy	Government
Production function	Explicit, detailed	Implicit, conceptual
Key data	Inputs, activities, outputs	Inputs, outputs and outcomes
Decision-making	Model-based, “scientific”	Political compromise
Objective	Maximize physical production (social welfare)	Improve efficiency of government spending
Perceived key challenges	Lack of computing power, data availability	Lack of enabling framework, limited incentives
Reasons for failure	Lack of incentives, inadequate feedback loops	Too early to say

There are obviously many views regarding the reason the Soviet system failed, but most analysts agree that the lack of incentives was one major factor. The inability of the system to process its own results and to use these results to adjust plans and objectives is another important feature. The classical example is the country with a 5 % annual growth objective, where the finance minister after four years of  3 % growth declares (with a straight face) that the country will have to grow 14 % in the last year of the 5-year program.

Which Form of Performance Budgeting Do Countries Actually Implement?

 

A number of countries around the world, including many developing countries, are attempting to improve the performance of their government sectors, and performance budgeting is often seen as an important aspect of these efforts. This is a very worthwhile objective, but one that often is difficult to achieve. One key question is: are countries primarily (and successfully) pursuing modern forms of performance budgeting, or are the reforms more similar to the type of program budgeting that was introduced in the 1960s?

The results are obviously mixed. However, when reviewing performance reform efforts in different countries, the following weaknesses are quite common:

- Many reform efforts are quite mechanical and ritualistic. There is little attention given to the quality of budget programs and indicators.

- Program and indicator definitions tend to focus on inputs and activities, and to some extent outputs. In many countries there are few examples of well-defined programs with clear outcome objectives and coherent indicators.

- Countries often retain detailed line-item budgets, effectively giving line managers little authority to manage their budget to meet objectives as efficiently as possible.

- Other rigidities are also prevalent, including on staffing decisions such as hiring and firing.

- Promotions and pay often continues to be based on seniority, giving very little financial incentives for good performers.

- Performance tends to have little or no influence on budget decisions. In many cases performance information is not even available or presented in a coherent manner when budget decisions are taken.

Conclusion

While many countries attempt to introduce modern forms of performance budgeting, as pioneered in countries like New Zealand, Australia and the UK, there are indications that many of them end up with schemes that are more similar to the failed program budget efforts in OECD countries during the 1960s. If this is correct, there may be good reasons to rethink and adjust ongoing performance budget reform efforts in these countries.

Comments

[Richard Allen]

Nice piece! Let's have more of the same. I agree with your general analysis. Another aspect of the 1960s-style budgeting that fits the Soviet mode was input-output analysis as pioneered by Nobel-prize winning Wassily Leontief. Highly regarded at the time, it fell into the same trap as program budgeting - over complex, driven by technology, simplistic equations, and computer power. You may be right that modern-day performance budgeting will suffer a similar fate, particularly in countries where the national planning mentality relations very strong (a connection you don't discuss in your piece).

[Sanjay Vani]

Agree with Allen - a nice piece! It is good to get some provocative titles and articles for a blog - just to generate discussion and debate!

In most countries that I had an opportunity to work and which experimented with the performance budgeting, the thing that struck me was not the lack of intellectual clarity on the concept of performance budgeting (a few people in the MoF/MoE had read and absorbed NZ and other thoroughly) but the absolute lack of systems to capture and process performance related information. Most of these countries have had made some progress in implementing state-of-the-art IFMS system but lacked proper systems to capture performance data. So what you see is performance budgeting but no performance accounting - the end result in such situations is apparent.

[Holger van Eden]

This made exciting reading but in my humble opinion program budgeting, and its implementation and use, have enough complexities to make it a challenging PFM reform all by itself. The comparison to "Soviet-style central planning" is a bit over the top for a number of reasons, but to be rhetorical: "where are the breadlines that program budgeting has caused". At worst PB seems to be ineffective and imply certain, but limited, data collection costs. That is not the same kind of societal stagnation one can attribute to central planning. There are, by the way, numerous processes and institutions, that still have some similarities to "Soviet-style central planning" - you might be working in one of those institutions! - this does not make them quite as destructive as the former Soviet system. I would also suggest that most of the Western European program budgeting initiatives in the sixties had their origin in US PPBS initiatives first implemented by the US Defense Department. It is interesting to note that many Eastern European counterparts that I have met were in the past quite surprised to learn that market economies like the US and other OECD countries actually used strategic planning in government quite extensively. I also think that program budgeting can be considerably strengthened if it is linked to periodic strategic planning exercises, either at the national or at the sectoral level. Planning institutions are something many transition countries are now rebuilding after initially burning them down fully after the yoke of the Central Plan was removed, and this is something I think we should support.

The differences between central planning and program budgeting are many, but let me just name the most prominent. Program budgeting tries to introduce rational incentives for government activities and decision-making where in the past there were none. Central planning on the other hand tried to take away incentives, because it saw them as unnecessary and wasteful. Why would one want to allow profits to be reaped, etc.. The other main difference is that central planning covered all economic activity, while program budgeting covers only government activities, i.e. those activities where the price mechanism cannot operate or not fully. Structuring those government activities appropriately in a program classification forms the basis of program budgeting, in my view, as it links the policy agenda to expenditure in an orderly way, not the myriad of program indicators that are sometimes developed in tandem. The main developmental question for PB is perhaps how to manage and use performance indicators for various decision-makers in the government hierarchy.

I would agree though that most PB systems are a tad overly-complex, if that is really what your saying...;))

[ahrainey]

Unfortunately people do not realize that there are three governments in the U.S. - federal, state, and local. For those in state and local finance, any comparison with the Soviet Union is an indication of a lack of understanding of the differences in governmental finance.

The federal government in U.S. does not have to have a balanced budget. The federal government in the U.S. does not have to work with rating agencies. State and local governments have to have balanced budgets and when they issue debt, they have to meet the requirements of the rating agencies utilizing such mechanisms as the "Financial Trends Monitoring System" of the International Cities and Counties Management Association (ICMA).

The author also does not recognize that in the U.S. governments must utilize Government Generally Accepted Accounting Principles (GAAP). The government GAAP for the federal government is promulgated by the Federal Accounting Standards Advisory Board (FASAB) while state and local government GAAP is promulgated by the Governmental Accounting Standards Board (GASB). Here again, no central planning allowed for those who understand accounting.

Finally, the author does not understand that in the U.S. state and local governments have been using performance based budgeting (pbb - no one uses program budgeting except theorists) in the form of strategic benchmarking, performance benchmarking, process benchmarking (through the GASB's Service Efforts and Accomplishments indicators), Kaplan and Norton's Balanced Scorecard, Activity Based Costing and Activity Based Management. Sorry, but no central planning here.

Perhaps it would be a good idea for the author to examine Minnesota Milestones or the Oregon Benchmarks or the STAT systems that Baltimore and New York City have used. It would also be a good idea for you to establish a dialogue with the ICMA and the Government Finance Officers Association (GFOA)who provide courses and publish articles on how city and state governments in the U.S. utilize "performance" indicators. There can be no centralized planning because in state and local governments democracy requires this concept called "full disclosure" and many states, like Oregon, have "budget laws" requiring that budgets utilize citizen budget committees.

The author is out of touch with reality. U.S. states and cities and counties are a different animal than the federal government and such comparison to "Soviet style central planning" shows that, like the folks in DC, the author is out of touch with main street. We (state, city, state) have been doing this for years. Rather than writing on theory, come play ball with us who do this for a living to see how it is practiced. I think that we could learn a lot from you and you would learn about the way most governments at the state and local levels deal with budgeting. Other organizations like the National Association of State Budget Officials (NASBO) and the Association of Government Accountants (AGA) would also be able to share some insight.

Anthony H. Rainey