More competition, and a wall just under the threshold

Below the public-tender threshold, a public body may award a contract by gré à gré; from the threshold up, it must publish a call for tenders. Provincial bodies award 4.5 times as many gré à gré contracts just under the threshold as just above it, and municipal bodies 3.3 times as many. At a round amount with no legal meaning, the same ratio is only 1.2.

The wall follows the law. Each time the threshold moved, the wall moved with it, including in 2022, when municipal bodies adopted the new threshold nine months after provincial ones. In the zone just under the threshold there are 36% more gré à gré contracts than the surrounding trend predicts.

Above the threshold, bodies cite an exception the law provides: 16% of gré à gré contracts there cite that only one contractor is possible, against 0.7% just below, where the most common ground is an amount below the threshold (39%).

A gré à gré contract awarded below the threshold is lawful. This finding describes the rule's effect on the market; it says nothing about the reasons behind any one contract or any one body.

An open public call closed in 2025 drew 4.2 bidders on average, against 3.2 in 2022. The share of single-bid calls went from 22% to 17%. In construction it halved, from 17% to 8.4%.

It is not a composition effect: many bodies started publishing in SEAO over the period, but among the 240 bodies that published every year, the decline is almost the same. The series starts in 2022: before then, the published bidder list does not measure the same thing.

When a single bidder answers a construction call, the contract exceeds the buyer's estimate in 57% of cases. With six bidders or more, in 20%. The median contract goes from 105% to 83% of the estimate.

The same slope holds in services and goods. It is an association, not a demonstrated cause: a call that draws a single bidder may also be harder, more urgent or poorly estimated.

Among construction contracts from an open call, published up to 2022 and observed for at least four years, 51% end with a final value above the award, usually by a little: median at 107% of the award. In services it happens less often (39%) but by more: median at 147%.

A higher final value is not necessarily an overrun: it includes changes the body asked for. SEAO publishes a reason for each change, with no amount, and for 21% of them that reason is only an amendment or change-order number.

More recent contracts are not read: a change often arrives years after signature, and a recent contract would wrongly look unchanged.

In 2025, 62% of awarded processes published in SEAO were awarded by gré à gré, but they account for only 26% of the awarded value. Since 2021 the share has ranged between 62% and 70%: it is not rising.

Gré à gré is provided for by law below the public-tender threshold and, above it, for exceptions such as urgency or a single possible supplier. Among the 419 bodies that published every year, the share is of the same order.

One health body in two gives at least 85% of its contracts by gré à gré; one school service centre in two, at most 44%.

Within one type, the gaps are wide. Among municipalities, one in ten gives less than 21% of its contracts by gré à gré, and one in ten more than 85%.

The same gap just under the public-tender threshold. Of its gré à gré contracts near the threshold, the median municipality places 10% in the last ten percent under it; one municipality in ten places 20% or more. A gré à gré contract awarded below the threshold is lawful: the gap describes ways of buying, not any body's conduct.

The single bid varies less from one type to another: its median is 17% among municipalities as among health bodies (17%).

No body is named. Each point in the charts is a public body, without its name, and a type of buyer is shown only when it counts enough bodies that no point singles one out.

Bid prices are published for 86% of awarded open calls, but for only 70% in services. The buyer's estimate is published more and more often: 37% of calls in 2024, 44% in 2025.

The weak point is elsewhere: for 21% of contract changes, the only published reason is an amendment or change-order number.

The median contract awarded in 2025 was worth $69K. From 2023 to 2025, between $36.1B and $41.2B was awarded each year, not counting the ten largest contracts, which weigh up to 20% of a year.

Awarded is not spent: a framework agreement records its ceiling, which may never be reached. These amounts are an order of magnitude, not a trend: from one year to the next, a few very large contracts are enough to move the total.