Lancaster Pollard Investment Advisory Group conducted a poll in February to find out how organizations were budgeting. The results are summarized below.

There is no “right” answer to budgeting for variable debt because the future paths of interest and tax rates are unknown. So what might be a better way?

Defining “Better”

Most organizations would say that a positive variance -- actual rates lower than budgeted rates -- is acceptable, but a negative variance is a problem. This could tempt an organization into budgeting at high rates. This strategy, however, runs afoul of the goal of maximizing the money available for the mission.

The better method is that which:

1) minimizes the probability of negative variance;
2) maximizes the amount of money available to the mission;
3) reflects a reasonable cost of capital to price services.

A Better Methodology

This approach is a variation on budgeting using the last rate plus some spread. It looks at the most recent rate, adjusted for calendar effects, which are described below.

R_normal = R_actual -D_april
where the variables are defined as:

R_normal: The rate adjusted for seasonal factors
R_actual: The most recent interest rate observed.
D_april: A seasonal adjustment for tax season. If the date of the most recent rate is between April 15 and May 15, then D_april = 0.80, otherwise D_april = 0

Once you have a normalized, or seasonally adjusted rate, the remaining step is to add a spread to that rate to reduce the probability that the rate will be exceeded (P_spread). The table below shows different spreads and probabilities.

Probability P_spread

Budget Rate =R_normal+ P_spread
Budget Rate = 2.90 + 1.03
Budget Rate = 3.93%

Lancaster Pollard Investment Advisory Group recommends using the lower probability spread. The reason is that interest rate moves are highly correlated with the actions of the Federal Reserve. It is rare for the Federal Reserve to move in more than 0.50% increments in tightening. The only exception to that pattern was in the early 1980s. Additionally, the tax-exempt rate change is typically 70% of the taxable rate change.
The implication is that it is rare that an actual rate above the probability rate will occur early in the budget year. This means that early in the year, there should be a positive budget variance. If the actual rate does rise above the budgeted rate later in the budget year, a negative variance, there should be a cushion from the prior positive variance.

Financial Discipline

The assumption embedded in the strategy is that management and the board have fiscal controls and discipline. If they see a positive variance and spend it early in the year, then there will be nothing later in the year. For those organizations with weak discipline, Lancaster Pollard Investment Advisory Group recommends establishing a restricted account, or interest rate reserve. Budgeted interest, credit enhancement costs and remarketing fees are deposited into this account monthly. Actual costs are then paid from this account, allowing surplus to build.

How well would the strategy have worked historically? The chart below shows the forecasted rate sets and the realized rates.

The red line shows how rates have moved on a weekly basis since the end of 2004, while the yellow line shows an estimate based on a December budget with the rate set 45 days prior to the end of the year. The blue line is based on a fiscal year ending June 30.

Conclusion
It is possible to minimize the risk of “negative surprise” and accurately reflect the cost of capital by using an interest rate forecasting model that begins with the last interest rate, adjusted for the tax season (if necessary) and adds a probability-adjusted spread. We believe it to be a best practice for budgeting.

Print this article