4 Ways to Make Your Revenue Estimates Less Wrong

Every year, elected officials decide how much to spend on which government programs and services by approving a balanced budget. But how do they know how much money they have to spend? Dust off your crystal ball. Setup the dartboard. It’s time to estimate revenues. And there is one thing we can say for certain about any revenue forecast: It will be wrong. You won’t be 100% accurate. But here are four ways to make your revenue estimates less wrong.

1. Know your revenues and how they work.

This may seem obvious, but your revenues are more than just dollar amounts that show up in the bank and in a financial report. They come from residents, visitors, businesses, and other governments based on what happens in the world around you. Before you can accurately estimate your revenues, you need to know what they are and how they work. Start by cataloging all your significant revenues into a revenue manual. Name. Description. Legal authority. Who pays them. Which department manages them. Exemptions. Account codes. Recent history of actuals. You can build your revenue manual as a document or a data sheet.

Your revenues might include taxes, inter-governmental aid, departmental fees and fines, and more. They all work differently, but most share the same basic formula: Price x Units = Revenue. Just like a business selling widgets! The price can be the property tax rate, a parking ticket fine, or even your share of state aid. The units are the total of the thing generating the revenue, whether it is taxable property value, the number of parking tickets, or total aid in the state budget. Make sure to identify the price and units for each item in your revenue manual.

2. Find data to analyze your prior year revenues.

Now that you know your revenues and how they work, you can start by explaining the amount collected in prior years. If a specific revenue grew, was it because the price changed or because the units increased? Both? With additional data, you can answer these questions! Use whatever data sources you have available to quantify the historical price and units for each revenue, and don’t forget to engage subject matter experts to help explain what you find. You may need to adjust for extraneous or one-time activity, but when you add it all up, you should get close to validating the revenue collected, or at least the trend, from its base parts. If not, you may still be missing a piece of the puzzle.

For example, let’s say parking enforcement revenue grew 10% last year. But you know that parking fines (the price) were increased from $20 to $25 last year, a 25% increase. Assuming the higher fines were in effect for the entire year, why did revenue grow less than the fines? The number of violations (the units) must have decreased (or at least the number of people who paid). The parking department can tell you how many tickets they issued for which violations and at which dollar amounts. It’s possible the higher fines deterred more parking violations. That’s good! Check if parking meter revenue went up since the two are related. Look at historical violations for the natural variability and scrutinize outliers. If violations last year were well below average, you may be seeing a new normal, which could be the new base for your forecast. Factoring in multiple fine categories, late fees, historical collection rates (tickets paid vs. issued) will add more layers of complexity.

To be fair, you probably don’t need a deep dive into every revenue your government collects. For smaller static revenues, reviewing the prior years’ total collections and projecting a simple straight-line trend forward is a reasonable and low risk approach. However, for larger dynamic revenues comprising the lion’s share of your budget, it is important to understand their base parts and what drives their past and future behavior. An oversimplification could mean big risk for your budget.

3. Make informed and reasonable forecast assumptions.

Now that you’ve decoded prior year revenues down to their base parts, here comes the hard part. You need to predict the future! You’ll need to make assumptions about what will drive those base parts (price and units) going forward.  History can be a guide, but you must also examine your current and expected future environment. Revenues may be affected by economic variables, legal limitations, population, traffic, crime, competition, politics, and departmental operations.

For example, Detroit levies local income taxes: 2.4% on residents and 1.2% on nonresidents commuting into the city. You know how much the city collected in prior years, but how do you predict the future? If you assume the tax rates (the price) aren’t going to change, then you need to make assumptions about future changes in employment and wages (the units). Those are two complex economic variables! Just kidding, those are four complex variables because there are two different tax bases and tax rates: residents and nonresidents. They may not move in lockstep yielding significant differences in revenue trends.

If you don’t have the luxury of an army of economists on staff to forecast variables like local employment and wages, consider seeking some outside help. In Detroit, we started an ongoing economic analysis partnership with our universities. When in doubt, keep your assumptions conservative and disclose what you know and don’t know. There will never be perfect answers! Do your best and consult experts. If your revenue estimates are tough to explain to stakeholders or don’t seem to make sense at a high level, take time to reassess. You may be overcomplicating it. Don’t let your methodology trap you. Be willing to walk away from a forecast that doesn’t add up.

4. Learn from your mistakes.

Remember, we know one thing for certain. When the final revenues come in, your estimates will be proven wrong! It’s just a question of how wrong, which is another way of saying “how right.” If you’re lucky, all the positive and negative variances will cancel each other out, and you’ll be praised for your forecasting genius. But the work doesn’t stop there! Examine where you went wrong. Did you miss some piece of the revenue puzzle? Did data availability or quality limit your analysis? Did reality turn out differently than your assumptions? Don’t be afraid of forecast errors. Take what you learn from them and build it into future forecasts. As long as the revenue estimates continue to improve and you can explain the results, your stakeholders will continue to have confidence in the process and your budget will stay on a sustainable course.