Election-timed spending
A Socratic walk-through of election-timed spending — reasoned out one step at a time, not lectured.
The question we started with
THE QUESTION #Why do public projects break ground in the year before an election rather than after?
Road resurfacing, a new clinic, a bridge repair, a school extension: all of them are useful in any year. Yet if you plot when the shovels actually go in the ground across a lot of jurisdictions, the answer is not "evenly." Spending — especially the visible, ribbon-cutting kind — tends to bunch in the months before an election and thin out in the year after it.
The cynical reading is bribery with public money, and sometimes it is. But that explanation is too easy, because it does not tell you why the timing is so regular. Bribery could be spread out. Something is making the calendar itself matter. What?
Reasoning it through
REASONING #Start on the voter's side rather than the politician's, and ask an innocent question: when you judge an incumbent, how much of the term do you actually weigh?
Honestly, not much of it. Political scientists studying retrospective voting have found repeatedly that outcomes in the final year carry far more weight than earlier years — voters behave as though the recent past is a better sample of an incumbent's quality than the distant past, and the distant past fades quickly from view. Whether that is rational updating or plain recency is argued about, but the empirical weighting is well established.
Now take that as given and reason forward. If the same project produces a lot of electoral credit when it is visible in year four and almost none when it is visible in year one, then the value of a project depends on when it is scheduled, even though its social benefit does not. And what does an actor do when an identical action pays differently at different times? Concentrate it where the payoff is.
Notice that this argument does not require anyone to be corrupt. An incumbent who sincerely believes she is the better candidate, and who wants to keep governing, has a reason to put the visible work where it will be seen — and even one who has no such intention will be outperformed by rivals who do. The incentive survives good faith.
Next, ask what exactly gets moved, because budgets are not fully flexible. Salaries, pensions and debt service cannot be shifted around the calendar. What can be shifted is the discretionary and capital portion: when a tender is issued, when a contract is signed, when ground is broken. So the cycle shows up less as a change in the total and more as a change in composition and timing — studies of municipal budgets, notably in Colombia by Drazen and Eslava, find pre-election shifts toward the categories voters can see, sometimes with no rise in total spending at all. That is a subtler prediction than "they spend more," and it is the one that tends to hold up.
There is a mechanical piece too, which is easy to miss. A groundbreaking is schedulable; a completion is not. Construction overruns, so an official who wants a photograph before polling day is far safer scheduling the start than the finish. That alone would tilt the observed pattern toward pre-election starts even without any voter psychology.
One caution against overstating this. Brender and Drazen found that political budget cycles are strong in new democracies and much weaker in established ones, and — more awkwardly for the cynical story — that running deficits does not generally help incumbents in developed democracies and can cost them votes. So the incentive is real, the behaviour is measurable, and yet the strategy may not reliably work where voters have long experience of watching it done.
The analogy
THE ANALOGY #Think of a plant that flowers on a photoperiod. It has energy available all year, but it holds the display for the window when a pollinator is present, because a flower opened in the wrong month is metabolically identical and reproductively worthless. The energy did not change; the timing was selected by an external rhythm the plant did not set.
the plant is entrained by a signal it cannot manipulate, whereas an incumbent often controls both the flowering and some of the calendar — in parliamentary systems a government can call the election itself, which is the reverse of the relationship and has no botanical counterpart.
Clarifying the model
THE MODEL #Three corrections keep this from becoming a conspiracy story.
First, most of the bunching is legal and visible. Budget documents, tender notices and capital plans are public. This is a pattern in the timing of ordinary decisions, not a hidden scheme, which is exactly why it can be measured at all.
Second, not every pre-election opening is strategic. Terms are four or five years, construction takes two to four, so projects authorised early in a term naturally mature late in it. Separating that innocent maturation from deliberate scheduling is the hard empirical problem in this literature, and studies that fail to do it overstate the effect.
Third, the cost of the cycle is not mainly the money spent — it is the sequencing. A project accelerated to meet a date, or slowed to avoid landing in the wrong year, is not being sequenced by need. Fixed election dates were meant to reduce the manipulable part of the rhythm, but they arguably sharpen this half of it: everyone now knows precisely which year is the visible one.
A picture of it
THE PICTURE #How to readthe loop is the term, and the arrow back from the fourth year to the first is the crucial one — costs are placed where they will be forgotten and benefits where they will be remembered, and the whole arrangement resets the moment the election is over.
What became clearer
WHAT CLEARED #Election-timed spending is best understood as an entrainment problem rather than a morality one. A recurring, publicly known date, combined with voters who weight recent outcomes heavily, converts an otherwise timing-neutral decision into one with a strongly preferred phase. Once the payoff has a phase, everything schedulable drifts into it — and it does so whether or not any particular official is acting in bad faith. Fix the rhythm and you fix the drift; the money was never really the variable.
Where to go next
ONWARD #- Whether fixed-term election laws dampen the cycle or simply make it more precisely aimed.
- Multi-year capital budgeting and independent infrastructure bodies as attempts to move sequencing outside the electoral clock.
- The same phase logic elsewhere: quarterly earnings management by firms, and end-of-fiscal-year spending surges in agencies that cannot carry funds forward.
Key terms
TERMS #| Term | What it means |
|---|---|
| Political budget cycle | the observed tendency for fiscal choices, especially their composition and timing, to shift systematically in the run-up to an election. |
| Retrospective voting | judging an incumbent by outcomes already observed rather than by promises, with recent outcomes weighted most heavily. |
| Discretionary and capital spending | the portion of a budget whose timing an administration can actually move, as opposed to salaries, transfers and debt service. |
| Entrainment | the alignment of an internal cycle to an external periodic signal, here the fixed rhythm of the electoral calendar. |
Every term the collection defines is gathered in the glossary.