Section 1
The two facts that cannot both be satisfied
Start with the money going in. In Sweden, the ROT deduction, the tax break that refunds part of the labor cost of home renovation, was temporarily raised from 30 percent to 50 percent for work paid between 12 May and 31 December 2025, a deliberate move to pull renovation activity forward. In Denmark, a renewed green craftsman deduction, the håndværkerfradrag, runs from 2025 through 2027, offering a deduction of DKK 8,600 per person in a household, up to DKK 17,200 for a couple, aimed at climate work like insulation, windows, and solar, and pointedly covering labor cost only, not materials. Both instruments are, by design, demand generators. They make hiring a tradesperson cheaper for the customer, which is the same as making the tradesperson's time more valuable. Now the floor. Nordregio, the Nordic Council's own research body, documents in its 2024 work on rural labour shortages and in the State of the Nordic Region 2024 that construction across the region struggles to fill positions, with vocational trades among the hardest to recruit. In Åland, plumbers, electrical and refrigeration technicians, roofers, and construction workers are all named as difficult to recruit. In Norway, vocational building and construction roles are flagged as shortage areas, made worse by a second-homes boom driving rural construction demand. In Finland, construction near urban areas is short of skilled labour. These are not gaps waiting for a wage signal. They are the supply floor. Put the two facts side by side. The state is paying to increase the quantity of renovation demanded. The quantity of people who can supply renovation is fixed at the floor. Both facts are real, and they cannot both be satisfied. Something has to give, and the thing that gives is the customer who does not get served, the price that rises without adding a single worker, and eventually the operator who staffed up for a demand curve that was always going to be withdrawn.
Section 2
Model one: the threshold, and the subsidy cliff
A subsidy with an end date is a threshold event wearing the costume of a gradual policy. Threshold models are built for exactly this: systems that behave one way, then flip sharply at a critical point. The Nordic subsidies have two thresholds hiding in them, one on the way up and one on the way down. The upward threshold is the one operators feel first. Below a certain level of subsidized demand, you absorb the extra work by stretching the schedule and pushing your best people. Above it, you cross into a regime where demand so exceeds your fixed capacity that the binding constraint is no longer sales, it is hours in the day. At that point the subsidy stops flowing to customers as intended and starts flowing to whoever holds the scarce labor, as higher prices. This is the quiet result: a labor-cost subsidy in a supply-constrained market is captured by the constrained factor. The refund meant to help the homeowner substantially becomes a price increase paid to the trade, because the trade can charge more when everyone has a coupon and nobody can add a worker. The downward threshold is the dangerous one, and it has a name worth using: the subsidy cliff. Sweden's enhanced ROT rate was written to expire on 31 December 2025. A temporary subsidy does not taper. It ends on a date. The demand it pulled forward does not return gently to trend. It falls off a cliff, because much of the work was not new work, it was future work brought forward by people racing to beat the deadline. When the deadline passes, you have already done next year's jobs this year. The order book that felt like structural prosperity was a pull-forward, and the cliff arrives on schedule. The operator lesson from threshold thinking is to distinguish structural demand from pulled-forward demand while you are still inside the boom, because they feel identical from behind a full calendar. The tell is the deadline. Work that clusters against a subsidy expiry date is borrowed from the future. Work that would have happened anyway is yours to keep. If you cannot tell which is which, you will over-read your own order book and staff for a demand level that has an expiry date printed on it. Threshold (the Nordic subsidy cliff), the tipping lens • Assumes: the subsidy has a critical activation level and a hard end date, and demand behaves discontinuously around both. • Fits because: a dated, temporary subsidy is a textbook tipping mechanism, up and then down. • Breaks when: the subsidy is quietly extended or made permanent, in which case there is no cliff and threshold thinking scares you out of investment you should have made. • Counteracts: the reflex to read a full order book as structural, durable demand. • May reinforce: excess caution that has you refuse growth that was actually going to last.
Section 3
Model two: comparative statics, and where a labor subsidy actually lands
Threshold thinking tells you when the market flips. Comparative statics tells you who captures the money in between, and the answer is counterintuitive enough to be worth deriving. Move one variable: a subsidy that lowers the effective price of tradesperson labor to the customer. In a normal market with an upward-sloping labor supply, the subsidy raises the quantity of work done. Some of the benefit reaches customers as cheaper renovation, some reaches workers as higher wages, and the market grows. Standard, healthy, intended. Now make labor supply vertical, which is what the Nordic floor means. The subsidy still lowers the price to the customer, so demand rises. But quantity of work cannot rise, because there is no one to do the extra work. The only variable left free to move is price. So the price of the labor rises until it chokes off exactly the amount of extra demand the subsidy created, and the market ends up doing the same quantity of work as before, at a higher price, with the government's money absorbed into that higher price. The subsidy did not buy more renovation. It bought the same renovation at a higher price, and the incidence, the question of who actually pockets the subsidy, lands on the scarce factor: the tradesperson and the firm holding them, not the homeowner the policy meant to help. This is not a criticism of the operator who takes the price. It is the rational response to your own scarcity, and you should take it. It is a warning about misreading where you sit. If your revenue rose during the subsidy, some of that rise is not you winning a bigger market. It is you capturing a transfer that ends when the transfer ends. Comparative statics tells you to separate the two on your own books, because one is durable and one has a cliff date. Comparative statics (price does not equal supply), the incidence lens • Assumes: you can move the subsidy variable alone and read the equilibrium shift; supply shape is knowable. • Fits because: the whole question is who captures a price subsidy when one side of the market cannot expand. • Breaks when: supply is only steep, not vertical. Real slack exists at the margin (retirees returning, cross-border workers from lower-wage regions), so the subsidy does buy some genuine extra output, and treating supply as perfectly fixed overstates the pure price effect. • Counteracts: the belief that subsidy revenue is the same as market growth. • May reinforce: cynicism that dismisses all policy as pure transfer, missing the real if modest supply response.
Section 4
Model three: game theory, cross-border poaching under a subsidy
There is a strategic layer the first two models miss. A subsidy that raises the price of Nordic tradesperson time does not just get captured domestically. It sets up a poaching game across borders and across firms. The players are Nordic firms, the funded consolidators entering the trades, and, critically, workers who can move. The subsidy raises the reward for holding scarce labor, which raises what every player will pay to hold it, which turns a domestic shortage into a bidding contest that can pull workers across borders from lower-wage regions. That is a genuine, if partial, relief valve on the supply floor, and it is why the floor is steep rather than perfectly vertical. But the same game has a trap. The player who can pay the most is not the operator with the best job economics. It is the player with a reason to pay above job economics, which, as in every trades market now, is often the consolidator buying capacity to protect an acquisition. During a subsidy boom, that player looks unbeatable, because it can outbid you for the exact crew you need to serve the exact demand the government created. You lose the labor race precisely when demand is highest, which feels like the market punishing you for winning. Then the subsidy reverts, the poacher's math sours, and the crew it overpaid for becomes a cost it cannot carry. The game does not reward the highest bidder. It rewards the player still standing after the cliff. Game theory (poaching), the strategic lens • Assumes: identifiable players bidding for movable, scarce workers with different payoff functions. • Fits because: a subsidy-inflated price for scarce labor is exactly what triggers a cross-firm, cross-border bidding contest. • Breaks when: labor cannot actually move (licensing barriers, language, housing), in which case the poaching game is thinner than it looks and the floor is harder. • Counteracts: the assumption that losing the labor race during a boom means your economics are wrong. • May reinforce: an arms-race mentality that has you overpay into a demand level with an expiry date.
Section 5
The structure-break flag
Now the flag that governs all three models, and the reason this case matters beyond the Nordics. Every model above assumes you can treat demand and supply as separate levers moving against a stable backdrop. The structure breaks when the two forces at the heart of this piece meet at the same moment: the subsidy reverts while the supply pool is still at its absolute floor. This is the specific whiplash. On the way up, demand was inflated by policy and supply was stuck, so price rose and everyone felt busy. On the way down, demand collapses on the cliff date while supply is still stuck at the floor, because the floor is a decade-long training problem that a two-year subsidy never touched. You get the worst of both regimes in sequence: you cannot serve the boom, then you cannot shed the cost fast enough for the bust. The people you fought to hire are still on the books when the demand that justified them expires by statute. Name this on your own calendar. The structure-break date is the subsidy expiry date. That is the day your boom-era model stops describing your market, and the day the question flips from how do I staff this to how do I survive the reversion without the crew I overpaid for.
Section 6
The proposed solution: GEER, then RADAR, then CHAIN
GEER: the lever order inside a subsidy boom The dominant channels here are demand-durability, mix, and cost-flexibility, not headcount. Rank the levers by reversibility. Pull first, cheap and reversible: raise price on subsidized work toward what the constrained market will bear, because if you do not capture the incidence, the subsidy simply lets your customer pay less while you do the same scarce work for the same money. Then tag every job as structural or deadline-driven, and quietly steer your fixed hours toward the structural work that outlives the subsidy. Pull second, more costly but still in your control: staff the boom with flexibility rather than permanence, using subcontract and temporary arrangements for the deadline-driven overflow so the cost can leave when the demand leaves. Resist the urge to convert every busy month into a permanent hire, because a permanent hire against temporary demand is how the cliff hurts you. Pull last, slow and expensive: fund genuine capacity through training only against the structural slice of demand you are confident survives the subsidy, never against the pulled-forward slice. This is the only lever that lifts the floor, and it must be sized to durable demand, not to the peak. The no-lever flag: nothing you do during a two-year subsidy adds a licensed craftsperson to the national pool on that timeline. The floor is set by training decisions made years ago. Any plan that assumes you can staff up to meet subsidized demand is a plan to hold cost you cannot shed. RADAR: a dated portfolio around the cliff Do now, near zero-regret: capture the price incidence, separate structural from deadline demand on your books, and hold your permanent headcount flat while meeting overflow with flexible labor. Hedge, cheap tail insurance: keep your fixed-cost base low enough to survive a demand fall to pre-subsidy levels, and write your subcontractor and temporary terms so they unwind cleanly on or before the expiry date. Defer with a pre-committed trigger: the decision to invest in permanent capacity or a training pipeline waits for a specific observable. Trigger it only if demand holds for two quarters past the subsidy expiry, proving it was structural and not borrowed. Until you see that, the irreversible move stays deferred, because the cliff is the whole risk. CHAIN: the base rate for demand subsidies into supply floors Match this to its reference class on structure, not surface. The class is temporary demand subsidies poured into constrained-supply markets: cash-for-clunkers style auto incentives, first-time homebuyer credits, pandemic-era home improvement booms. The base rate is consistent and unkind. Prices rise during the subsidy, activity spikes and then falls below trend after expiry because demand was pulled forward, and the businesses that expanded fixed capacity to meet the peak are the ones that struggle in the trough. The survivors treated the boom as temporary from day one. Tilt by present state: how large is the deadline cluster in your book, how flexible is your cost base, how exposed is your local pool to poaching. Then subtract the counterfactual: some Nordic renovation demand, especially the climate-driven insulation and solar work in Denmark's green deduction, reflects a durable long-term shift and would happen without the subsidy, so not all of it cliffs. The matrix-break flag: if a subsidy is quietly made permanent, the base rate from temporary programs stops applying, and you should re-run the analysis as structural demand.
Section 7
The blind spot
Here is what this ensemble cannot see. Every model treats the subsidy as an exogenous shock, something that happens to your market from outside. It is blind to policy path-dependence: the possibility that a temporary subsidy becomes politically impossible to remove, extended again and again because the trades and the households both organize to keep it, so the cliff that the models are built around never actually arrives. If that happens, the operators who stayed lean and refused to expand, exactly the behavior the framework recommends, will have left durable, subsidized demand on the table for years. The models are calibrated for reversion. They cannot tell you whether this particular subsidy is the one that becomes permanent, and that is a political forecast no economic model contains.
Section 8
The fitness test
Run this before you hire one more person to meet the boom. Look at your current order book and sort it into two piles: work that would exist at the old subsidy rate, and work that exists only because of the temporary enhancement and its deadline. Now ask: if the enhanced subsidy expired tomorrow, could you cover the crew you are about to hire from the first pile alone? If the answer is no, you are not staffing a growing business. You are staffing a demand curve the government will withdraw on a printed date, and the day it reverts, while the craftsman shortage sits exactly where it always was, is the day you find out whether you read your own order book or just enjoyed it.