Spending
A budget is a forecast, and forecasts are supposed to be wrong
Most household budgets collapse because they are treated as promises about the future rather than estimates that get corrected as evidence comes in.
By Aditya Ramaswamy3 min read
What the document actually is
A budget looks like a set of rules and is generally described as one, which is where the trouble starts. What it really contains is a set of predictions: how much this household will spend on each of several things over the next month, written before the month happened by someone who cannot see it.
Every forecast has an error attached. A weather forecast that is occasionally wrong is not a failed weather forecast, and a budget that is out by some margin in several categories is behaving exactly as a forecast behaves. Treating that as a moral failure is the reason so many people abandon the exercise in the second month.
The useful version is closer to an instrument than a contract. It exists to tell you the difference between what you expected and what occurred, because that gap is the only new information the month produced.
Too many categories buys precision you cannot use
The instinct when a budget goes wrong is to add detail — split food into groceries and eating out, then split groceries again, then separate household goods from cleaning products. Each division feels like an improvement in accuracy. Mostly it is an improvement in bookkeeping.
Two costs arrive with that detail. Every extra line takes effort to maintain and adds a judgement call about where an ambiguous purchase belongs, and ambiguity multiplies as the categories narrow. Meanwhile the extra precision only helps if you would act differently on a fine-grained number than a coarse one, which is rarely the case.
The other problem is that many small categories hide the total. Twelve lines each a little over budget look like twelve minor errors and feel survivable. Aggregated into one number they would have been an obvious overshoot, visible in week two rather than at the end.
Irregular costs are what actually break a month
Most budgets are built around the recurring monthly picture, and most of them are broken by things that do not arrive monthly. Annual renewals, servicing, replacements, travel, the dentist, gifts clustered into particular months — none of these are surprises in any real sense, since almost every household has some of them every year.
Handled month by month they look like bad luck. Handled as an annual total divided by twelve, they turn into an ordinary line that happens to be paid out in lumps. Suppose the year contains 2,400 of such costs: that is 200 a month, and the budget that includes 200 for it is simply a more accurate forecast than the one that does not.
The category itself is not the point. Naming the arithmetic is. A forecast that omits a class of predictable expense will be wrong in the same direction every year, and being wrong in the same direction repeatedly is the one kind of forecast error that is entirely fixable.
Variance is the output, not the failure
Once a budget is understood as a forecast, the interesting figure is not what was spent but the difference between the estimate and the outcome. A category consistently under is money that could be allocated somewhere it is needed. A category consistently over means the estimate was wrong, and the estimate is usually the thing that should move.
That reframing removes most of the guilt, which matters practically rather than emotionally. Guilt makes people stop looking, and a budget nobody looks at has no function whatsoever. A budget revised three times in a year because the numbers were wrong is working properly.
It also stops the common error of treating an underspend in one category as licence to overspend elsewhere without noticing. If the categories are few enough, that trade is visible and can be made deliberately, which is a considerably better outcome than making it by accident.
The version that survives a real month
A budget that lasts tends to have few categories, a realistic allowance for the irregular, and one summary number that gets checked often enough to be actionable. Whether that number is what remains for the month, or the amount set aside, depends on the household and on what it is trying to fix.
It also tends to be revised without ceremony. The first attempt is a guess made with no data. The third attempt has two months of evidence behind it and will be markedly better, which is the ordinary way any estimate improves.
What none of this settles is what a particular household should be spending or setting aside, since that depends on income, obligations, dependants and local costs. A budget is a tool for seeing your own numbers clearly; decisions of any consequence built on top of it are worth discussing with a regulated adviser.
Common questions
How many categories should a budget have?
Few enough that you can hold them in your head and still act on them. Many households do well with a handful covering housing, transport, food, commitments and everything else. The right number is the one you will still be maintaining in three months, which is almost always smaller than the one you design on the first day.
What if income varies month to month?
Forecasting the spending side becomes more important, not less, because the income side cannot be pinned down. A common approach is to budget against a conservative estimate of income and treat anything above it separately, though how well that works depends heavily on how variable the income actually is.
Is budgeting software better than a spreadsheet?
Both work, and the difference matters less than whether the thing gets used. Automatic import saves effort and reduces the chance of abandoning it; entering figures by hand makes you look at each one, which some people find is the entire benefit. Neither improves a forecast that ignores irregular costs.
Aditya covers spending, saving, debt and the questions readers actually send in and thinks most subjects are more interesting once you know how they work.
