The Star Tribune Editorial Board says that not counting inflation in the state's spending forecast is "a foolish accounting gimmick put into law years ago that has distorted this state's budgeting ever since. Inflation is real and must be counted when projecting future expenses." ("Economic forecast signals caution," March 3.)

As the person responsible for first taking inflation out of the spending forecast when I was Minnesota's commissioner of finance in 1990, I could not disagree more.

First, though, a clarification. The Editorial Board, like many others, has adopted as a mantra that since inflation is counted in the revenue part of the forecast it should be counted in the spending forecast as well. That would make sense if the premise were correct, but it is not.

The revenue forecast answers this question: "How much revenue can we expect the state to collect if we assume that the current tax laws remain in effect?"

To answer that question, we need to make estimates of how many people will be working, how much those people and the state's businesses will earn, how much they will spend and on what. We also need to estimate the value of business property. These estimates form the basis of the forecasts for the personal income, sales, corporate income and business property taxes — the taxes that fund most of state government.

In making these estimates, the expected rate of inflation is more or less irrelevant. For example, if people are going to have higher incomes because labor is in high demand, or because stock prices are rising rapidly and they can cash in their capital gains, we don't limit the revenue forecast to the rate of inflation. We actually estimate how much people and firms will earn — even if the estimated rate of growth is much higher, or much lower, than the rate of inflation.

In fact, revenue from the personal income tax grew at twice the rate of inflation in the 10 years from 2005 and 2015.

Once we estimate what people and businesses will earn, we can estimate what they will spend and on what. How much they spend depends a lot more on what they earn than on the rate of inflation. They cannot spend what they don't have.

So, the revenue estimate is designed to tell us how much revenue the state can expect to collect if we assume that the current tax laws remain in effect. It does not and should not tell us what people and businesses would earn and spend if everyone simply got an inflationary increase.

What about the spending forecast? How we handle inflation on the spending side depends on what question we want the forecast to answer about what the state buys with its money. There are two choices:

First, we could ask, much as we do with revenue: "How much would it cost the state to pay for what is currently required by law?" The answer to this question depends on what is in the law.

Interestingly, there are laws that say costs will increase at the rate of inflation — several of our entitlement programs operate that way. Many of those same programs also operate under laws requiring the state to spend based on the number of people served. Since the law says that costs will rise based on those factors, the forecast says so as well.

But for many other programs the law does not require the state to spend more — because of inflation or for any other reason. In those cases, spending beyond what the law allows is not included in the forecast. If the governor and Legislature want to buy more (or less) they would have to put that increase (or decrease) into the budget and pass it into law.

This is how the forecast works today.

Our other choice would be to answer the question: "How much would it cost the state to continue doing everything it has been doing in exactly the way it has been doing it?"

To answer this question, we would estimate inflationary and other cost increases for people, materials, travel, office space and all the other things that go into providing public services exactly the way we've been providing them, add in the cost of serving more people, then apply those increases to the budget. This "status quo plus inflation" approach is the one favored by the Star Tribune and many others.

There are strengths and weaknesses to each approach. Our current "what the law requires" approach has the advantage of linking the forecast to a fixed base — current law. It provides clarity as to whether a program is being cut or increased, since it either gets less or more than it would have under current law. Finally, this approach challenges programs to do more with the money they have. It pushes decisionmakers to decide whether they want to buy inflation in old programs, improve those programs to get more for their money, or buy new, higher-priority programs.

Critics of this approach argue that it is unrealistic to assume government programs won't cost more. They also argue that it makes the state's fiscal challenges look easier to solve than they really are, since the difference between in-law revenue and in-law spending (the so-called "surplus") is greater than the gap between in-law revenue and spending-plus-inflation.

Those who favor the "status quo plus inflation" approach see its greatest strength as portraying the state's budget challenges more realistically by recognizing that costs do in fact increase. Naysayers, including me, argue that this approach perpetuates the status quo and leads to paying more for what we are already getting, rather than getting more for what we are already paying.

The status quo plus approach even leads to the peculiar result of programs receiving more money than under current law but claiming to have been "cut" because they got less than their inflated forecast.

Whether and how to count inflation in the spending forecast involves a choice of what you want the state to buy with your money. If you want state decisionmakers to buy more of the status quo at a higher price, the Star Tribune has it right.

On the other hand, if you want the governor and Legislature to buy the most they can with the money they have, leave budget forecasting alone (and even consider removing many of the autopilot inflation requirements in current law).

Peter Hutchinson is a former Minnesota finance commissioner, Minneapolis school superintendent and gubernatorial candidate. In 1990, he presented the first state budget forecast without automatic inflation built into every program.