Government Performance: Lessons Learned

Lesson learned: corporate performance management is hard, government performance management is harder

Businesses have bottom lines: profit (or loss). Government do not. Governments cannot validate performance management objectives like businesses can. That’s not to say that performance management in business is easy. It’s just that the business bottom line can demonstrate when performance structures are not aligned. Especially when performance objectives, outputs and outcomes, are not met yet profit increases. Or, vice versa.

Workshop exercises at the 2019 FreeBalance International Steering Committee (FISC) in Cascais Portugal demonstrated this government performance problem. Government don’t run like business, as much as many would like to think they could. FISC is not a user group. It’s not a sales event. It’s an opportunity for FreeBalance to learn from governments, and for governments to learn from each other. Effective management of public finances is not a “zero-sum game.” Countries benefit when other countries improve fiscal stewardship.
Our approach this year was to create a progressive set of 7 workshops covering the modernization of Finance Ministries. As always, there were sessions and workshops where governments changed the FreeBalance product roadmap. My sense is that the workshops helped the senior public servants selected by their governments to attend better prioritize roadmap choices. Which is important because the roadmap has grown significantly in the past few years as we leverage the extensible characteristics of the FreeBalance Accountability Platform.

Lesson learned: integrate performance and financial metadata

Government financial structures, the “Chart of Accounts”, is often more complex that in private sector accounting. It’s quite common to see national government COAs with over 25 digits with 5 or more segments. The COA often includes budget, organizational, program, economic purpose and fund source data. The COA can be simplified by using side concepts to roll-up financial information to different structures like government goals, Sustainable Development Goals, Classification Of the Functions of Government, and Government Financial Statistics, among many others. This demonstrates spending against goals, but not the effectiveness of spending unless outputs and outcomes are linked directly to the COA through the “Chart of Goals”. Performance measures can be integrated with financial systems through integration of the COA with the Chart Of Goals where the COA has the budget & accounting classifications, while the COG links the COA with outcome and output goals at different aggregate levels.

Lesson learned: some private sector practices can help government effectiveness

Businesses and governments have difficulty distilling performance structures. We found this through an interesting workshop that engaged all participants in groups. One exercise looked at the Objectives and Key Results framework (OKRs) initially developed at Intel and used a Google. Participants found that the OKR rules helped eliminate dubious and unnecessary performance measurements. As described in the book, Measure What Matters by John Doerr, organizations can only improve through measurement, only if measuring what matters. Another lesson: successful government effectiveness improvements need to start at the top where senior officials work out their performance measures first.

Lessons learned: look for “positive deviance” rather than “best practices”

Country-led reform tends to be more effective than the use of so-called “best practices”. These practices were often developed under different contexts and circumstances. FISC discussions identified a number of home-grown practices that operates well in unique circumstances. In the agile country development methodology, Problem-Driven Iterative Adaptation (PDIA), these are known as positive deviance “where reforms have led to more effective solutions to public sector problems than is normal“. The solution to improved performance often comes scaling some existing practices and processes within organizations to ministries.

Lessons learned: there are multiple methods to improve performance through experimentation

It can take some time to improve government outcomes. For example, major educational programs are unlikely to witness results for two to five years. Nevertheless, there are ways to improving some outputs and outcomes through low-cost experimentation like design thinking methods, A/B testing feedback, surveys, efficiency statistics, and PDIA.

Lesson learned: evidence-based decisions need unified system design

Finance Ministers make important government decisions. Evidence-based decisions are more effective. The problem is that data from numerous financial subsystems and government statistics is often inconsistent or too late. There can be multiple versions of the truth because metadata is inconsistent, leading to poor decision support. Some governments try to overcome this problem through metadata management and data warehousing software. The results are often disappointing because of legacy systems that do not support modern integration methods leading to bad practice integration. Or, delays because of semi-automated integration.

Lesson learned: smart requires metadata and controls integration

Integration is more than a decision-making necessity. Separate information systems lead to controls inconsistency. Some software used in government does not support budgets or commitments. For example, a lack of integrated revenue, debt, and commitment visibility often leads to cash crisis, with many governments going into arrears.
The most common controls integration problem in countries is procurement because of numerous touch points with commitment accounting that are almost never integrated when using separate systems:

  • Tender in procurement system may not reflect intended budget use as expressed in Chart of Accounts and government goals as expressed in Chart of Goals for value for money
  • De-commitments, obligations, de-obligations, and multiple year commitments arising from the procurement process would need to be manually entered in commitment accounting opening up errors, non-compliant processes and opportunities for fraud
  • Lack of integration on goods received, goods returned and contracts could result in incorrect payments such as not reflecting progress payments, penalties, hold backs, acceptance testing periods, and fixed assets
  • Lack of integration of vendor information could lead to payment errors, inability to provide vendors with information about when payments expected, inability to get discounts because of manual systems, inability to negotiate quantity deals, and inability to combine payments or support partial payments

Budget preparation and budget execution integration have multiple points of control failure:

  • Errors in setting up budget controls and allotments in core financials can differ from the approved budgets
  • Over reliance on budget transfers and virements during budget execution often results in spending that doesn’t reflect original budget
  • Danger that transfers and virements  could break compliance procedures without integrated controls
  • Lack of of debt, development (including donor off-budget plans), capital, operating and salary budgets integration leads to inefficient budgets, reduced liquidity (poor predictability of debt repayments, increased borrowing), and lack of fiscal sustainability after capital acquisitions

Government payroll integration with commitments is another common problem given the impact of wage bills:

  • Lack of integration between HR & payroll systems makes it difficult to expose ghost employees, confirm that public servants have necessary qualifications, and calculate pensions accurately for some
  • Nonsalary public servant spending like training courses, travel costs, and expenses can lead to payroll that exceeds commitments
  • Political influence to increase salaries or bonuses can lead to exceeding commitments
  • Ministries can be out of compliance with salary scales
  • Poor trend predictability of salary expenses leading to exceeding budgets or underspending, particularly training, by end of fiscal years
  • Often inability to effectively accrue leave in countries running accrual accounting