Best Practices

The Use of Trend Data and Comparative Data for Financial Analysis

A government’s annual comprehensive financial report provides a wealth of data that the government itself and other interested parties can use to analyze the government’s financial health. Of course, considering financial data in isolation can lead to inappropriate conclusions--appropriate context is essential for sound interpretation. For state and local governments, trend data of the government itself and comparative data from other governments frequently have provided this context for evaluating current-year financial data.

The National Council on Governmental Accounting (and later, the Governmental Accounting Standards Board - GASB) saw value in the use of trend data for financial analysis as evidenced by their recommendation to include up to a dozen schedules of trend data in the statistical section of the annual comprehensive financial report. Likewise, the GASB has mandated the presentation of trend data for both defined benefit pension plans and public-entity risk pools. The Government Finance Officers Association (GFOA) also has recognized value in the use of comparative data from other governments as evidenced by its ongoing compilation and publication of a Financial Indicators Database that presents selected data from annual comprehensive financial reports submitted to GFOA’s Certificate of Achievement for Excellence in Financial Reporting Program.

GFOA makes the following recommendations for the benefit of government officials and other interested parties who wish to use financial data from the annual comprehensive financial report to analyze a government’s financial health:

  • GFOA believes that a government’s own past performance normally is the most relevant (but not exclusive) context for analyzing current-year financial data;
  • A government’s own experience typically is best expressed in the form of trend data for key financial indicators (e.g., revenues, expenditures, fund balance);
  • The usefulness of trend data often can be enhanced by examining the percentage relationship among data elements over time (e.g., local revenue as a percentage of total revenue; public safety expenditures as a percentage of total expenditures);
  • At a minimum, five years of data typically are necessary for effective trend analysis;
  • Conversely, trend information eventually loses relevance over time because of changes in circumstances. Accordingly, typically no more than ten years of data should be considered;
  • Items that potentially distort trends (e.g., one-time items or changes in underlying assumptions or structures) should be carefully noted;
  • When constant dollar presentations are used, it is recommended that the basis of the constant dollar adjustments be disclosed and the unadjusted (non-constant) data also be presented. This allows individuals to make their own decisions regarding whether and how price level adjustments should be made;
  • Appropriate comparisons of a government’s own data with the data of other similar governments also may be useful for purposes of financial analysis. However, care must be taken to ensure that such comparisons are valid. Considerations that affect the validity of data comparisons among governments include the following:
    • Are the governments of the same level (i.e., state, county, municipality) and type (e.g., general-purpose, special-purpose)?
    • Are there significant differences in the scope or quality of services provided?
    • Are there significant differences in the number of those served?
    • Do the governments define categories in the same way?
    • Are the governments from regions where costs and similar environmental factors are comparable?
    • If costs being compared include significant depreciation expense, were the capital assets being depreciated acquired at roughly the same time (i.e., to avoid the distortions inherent in historical-cost depreciation)?
  • Comparisons with other governments may be further enhanced by using trend data for these governments rather than relying exclusively upon current-year data.
  • Board approval date: Friday, February 28, 2003