Nonprofit Recovery Using Statistics and Business Skill

Nonprofit entities seeking financial turnaround have several tools at their disposal, including statistical techniques and sound business procedure. The word “nonprofit” does not imply a preference of deficit over surplus; it means that there are no owners, no dividends, and no one to profit financially from the venture. It’s only a distinction among organizations: Nonprofits should be run in a businesslike manner and have a surplus, and there is no better time to begin a recovery than budget season.

Budget season requires more than documentation of expected revenues and expenses in upcoming fiscal years, and recovery involves more than just developing typical spending plans. Building a budget can be done several ways. For example, upper management may use a top-down approach, while operational or program managers create a bottom-up budget. The tools should be suited to the size of the entity and the level of detail desired.

A mix of the bottom-up method with statistics provides a strong starting point. Using statistical techniques can be enlightening, can add a measure of confidence to forecasts, and is not difficult to implement. Most important, these techniques reduce variances (and can catch accounting irregularities, even embezzlement), and this means submitting fewer cases to the board each month.

To begin, acquire ample history on every revenue and expense. Obtain this information from prior financial reports or download it from accounting software in the form of an ASCII file to a spreadsheet. A minimum of three years of information is necessary to use trend analysis, the primary statistical technique discussed here.

Trend analysis for budgeting purposes is simply regression analysis using time periods. It’s as easy as looking at three years of revenue or expense history to gauge where the organization is going.

To illustrate, graph the organization’s last three years of budget history. With dollars on the vertical axis and years on the horizontal axis, fit (as closely as possible) a straight line through the three points on the graph (Exhibit 1). Now add a fourth year (2004) to the horizontal axis. Follow a vertical line straight from 2004 to where it intersects with a projection of the existing trend line: this is projected 2004 dollars.

A program manager will not know whether a nonprofit can actually bring in the revenue or afford the expenditure suggested by a trend analysis until a bottom-up budget is constructed. For example, if the revenue item plotted in 2004 for workshop participants is $30,000, does that equate to a manageable number of participants at a reasonable rate increase over 2003? If the $30,000 revenue trend is much lower than the budget built, a business plan is in order. A business-as-usual attitude will not suffice.

Using the trend graph puts the nonprofit in a much better position: Its knowledge about the budget has expanded. It can realize what is achievable and what is not from the use of statistics or trending. The analyis also highlights where to focus future management efforts.

Trend analysis for larger organizations requires more work. For example, to establish a trend analysis on three years of workshop revenue, this information must be regressed to the related time periods. Larger organizations can be even more sophisticated and use marketing efforts instead of time periods. If 2001’s revenue was $15,000, 2002’s revenue was $20,000, and 2003’s is projected to be $25,000, what would be the 2004 trend in revenues? For a small organization, this can be estimated from the linear progression. But for a larger organization, with hundreds or thousands of accounts, a different tool is needed.

For larger organizations, the power of a computer spreadsheet program is essential. Exhibit 2 shows revenues by year as it would appear in a Microsoft Excel spreadsheet. The trend formula is used to calculate the revenue for 2004. Excel’s “Trend” function returns values along a linear trend by fitting a straight line to the known values, using the least squares method. It works exactly like the trend analysis described above.

The Trend function returns a value of $30,000. Management can now anticipate the workshop manager submitting $30,000 in budgeted revenues for the year 2004, assuming that no extraneous factors come into play. But remember that this is linear trending, not curvilinear. (Excel provides other formulas and variations with higher levels of required expertise, such as polynomial trending.) A nonlinear revenue history, such as $15,000, $19,000, and $26,000, is more difficult to project confidently.

Using the Trend function, the estimate for 2004 would be $31,000. The RSQ function can be used to establish the strength of the trend. RSQ returns the square of the Pearson product moment correlation coefficient through the given data points. Setting up this formula requires only the first two components of the Trend formula (Exhibit 3).

The result is an RSQ of 0.98. An RSQ of 1.00 means that all the data points reside on a straight line; a 0.98 RSQ represents the variation from this straight line. At what RSQ value should the trend analysis be ignored? Several factors will contribute to the decision. Generally, an RSQ of 0.75 or higher is considered a strong trend. Anything less probably points to something unusual that affected performance: a strong economy, an accounting irregularity, a popular workshop leader. The closer the RSQ is to 1.00, the more confidence that the projection is accurate and the trend will continue.

With a 0.98 RSQ and a Trend result of $31,000 in workshop revenue, the acceptable range of a budget submission from a manager may vary, but an excellent guide is materiality. Materiality begins in the 5% to 10% variance ranges. Thus, the workshop manager could submit anything in the range of $29,450 to $32,550 and the submission wouldn’t be materially different from the Trend result of $31,000; any submission beyond this range requires an explanation.

Thus, with trend analysis and RSQ evaluation, the organization can manage the budget even as it is being constructed. With statistical confidence, budget submissions can be evaluated based on predetermined trends and strengths.

Trending can also provide a quick, easy global examination of an organization’s finances. Used annually, this exam is a good steward of the organization’s funds and helps ensure the success of the nonprofit’s mission and programs.

Using at least three years of history for every revenue and expense line item, one can calculate a trend for 2004. What conclusions can be drawn? Is 2004 a surplus or deficit? Is it materially different? Run an RSQ on the historical surplus or deficit figures. Is it a strong or weak trend?

If the trend in recent years is toward a deficit, there is a potential problem. Conversely, a progression to a surplus can indicate a healthy organization. Regardless, it is important to investigate and drill down to discover the reasons behind the trend. For a deficit, lagging program revenues, compounding salary expenses, uncontrollable expenses, and mature membership bases are frequent culprits. Other typical suspects are unsound business practices or insufficient internal controls, such as hard-copy timesheets requiring multiple verifications that are incompletely performed and would be better off automated.

Reporting each program’s operating margin is critical. Follow-up with corrective action is essential. Responsibilities should be tied to performance, customer satisfaction to caring, reporting to honesty, and camaraderie to respect—all will bind a nonprofit’s mission with success.

Business turnaround requires operational vision, business acumen, and the scrutiny of every line item. This author’s experience with the Tuckahoe Family YMCA had the following result: A material deficit was turned into a record surplus of $262,000. Implementing the changes necessary for the turnaround took a cooperative and dedicated staff, and a new business mentality. Consecutive surpluses of 4.3% of total revenue and 10% annual revenue growth are the results to be had—but only with time.

With trend analysis, budgeting can be taken a step further. Trends based on the stronger of three or four years of history diminish linear effects, as do polynomials that establish curvilinear trends. Build spreadsheets that provide options of the strongest trend (linear or curvilinear, based on varied years of history). Explore regressions to things other than time periods, such as events on which revenues depend. Expect to see physical plant improvements, better bond ratings, enhanced endowments and investment balances, valuable and dynamic new board members, and advances in information technology. The nonprofit’s recovery will be systemic and cyclical, and it will survive and flourish.

The CPA Journal is broadly recognized as an outstanding, technical-refereed publication aimed at public practitioners, management, educators, and other accounting professionals. It is edited by CPAs for CPAs. Our goal is to provide CPAs and other accounting professionals with the information and news to enable them to be successful accountants, managers, and executives in today's practice environments.