Layers

Built structures

New buildings and other structures produced through construction investment. This is an annual flow of new construction. The existing stock of buildings supports output across several layers.

What-if model · Chosen targets, conditional projections and calculated gaps.

  1. Estimate todayToday · 2026$54.2TStarting estimate
  2. Conditional projectionOn course · 2035$62.4T$8.2T projected growth
  3. Historical-benchmark comparisonBaseline + tier 1$71.1T$8.7T tier 1 gap
  4. Chosen scenarioChosen 2035 target$105.7T$34.6T tier 2 gap

Annual output at 2026 purchasing-power prices. The historical-benchmark comparison includes input adjustments and is limited to the portion used for this target; it is not a physical ceiling.

Where the map's three numbers come from

These amounts belong to built structures alone. The first is projected growth under the baseline assumptions. The other two are calculated gaps to a chosen target. They do not predict how much new output will actually arrive.

PartCalculationAmount
Baseline growth$62.4T − $54.2T$8.2T
Tier 1 gap$71.1T − $62.4T$8.7T
Tier 2 gap$105.7T − $71.1T$34.6T

Display values are rounded. The underlying calculation uses the full precision in the CSV.

Why is this the sector target?

The model first chooses a $448.5T world economy, then chooses what that economy contains. This layer receives 23.6% of that total: $105.7T.

For this layer, the spending-response factor is a default of 1.00: spending is treated as rising one for one with income in the allocation rule. This value was assigned outside the household-demand categories; it is not a measured forecast for this sector.

The calculation combines a macro baseline allocated at today's sector shares with an additional gap allocated using those response factors. It then scales all sector targets to the same world total. A different allocation rule gives different sector targets and tier gaps.

Source files · inputs/stage1-v3/requirements/target_economy.csv · inputs/stage1-v3/requirements/code/model.py · inputs/stage1-v3/requirements/report.md

Compare two target choices already in the source
Target mixThis sector's targetIts tier 2 gap
Selected spending mix$105.7T$34.6T
Every sector simply doubles$108.5T$37.4T

Both choices allocate the same total world target. This is the upstream v3 comparison that v4 inherits for its sector targets; it is not a new run of the v4 shortage model or a probability range.

Source files · inputs/stage1-v3/requirements/layers_v3.csv

Why the other construction levels are $62.4T and $71.1T

Step in the inherited calculationAnnual output
Trend-based baseline before the input adjustment$67.9T
Subtract the assumed baseline finance shortfall−$5.5T
Baseline counted by v4$62.4T
Historical-benchmark comparison before the workforce adjustment$74.2T
Subtract output attributed to the workforce constraint−$3.1T
Historical-benchmark comparison used for the tier split$71.1T

The historical comparison recorded in the source is world real construction value added 2003-2012 (UN AMA ISIC F): 3.54%/yr. The projected levels and adjustments depend on applying that history and the input assumptions to this scenario.

Source files · inputs/stage1-v3/requirements/layers_v3.csv · inputs/stage1-v3/repair/baseline/layers.csv · inputs/stage1-v3/repair/agent4_bottlenecks/blocked_by_layer.csv

A separate v4 demand test, conditional on doubled world income, gives $97.2T of construction demand against the chosen $105.7T target. Its $8.5T difference is a reason to question the target mix, not proof that this much construction will be lost.

Show the calculation

Today + on-course growth + tier 1 + tier 2 = target
$54.2T + $8.2T + $8.7T + $34.6T = $105.7T

Display values are rounded. The CSV retains the precision used in the checks. Each term carries its own evidence label.

ComponentAmountEvidence
today$54.2T[E] Estimated
growth$8.2T[A] Assumed
tier1$8.7T[A] Assumed
tier2$34.6T[A] Assumed
target$105.7T[E] Estimated

What constrains this chosen scenario?

Each row below calculates a constraint if this target mix is attempted. It is not a forecast of lost GDP. Each test treats the other inputs as available, so the amounts overlap. The model credits the largest individual block within each tier when calculating explained output.

Part of the gapMissingLargest shortage explainsUnexplained
Tier 1$8.7T$8.7T$0bn
Tier 2$34.6T$15.4T$19.2T
Read the model's explanation for each tier

Transformers and grid equipment on course meets 57% of need; at its fastest precedent (US generator step-up transformer demand x3.74 in 6 years (2019-25, Wood Mackenzie via pv m) it meets 99%: on course it grows at its trend, not its record pace. Buyers (repaired baseline demand test at the max-output income): UNTESTED: income elasticity is a default (non-household layer); price elasticity A or zero

Construction trades would have to grow 7.0% a year 2026-35 against 5.4% at its fastest precedent (1.3x faster)

Would the extra output be bought?

At its assumed 2035 price, the model estimates demand of $97.2T against a target of $105.7T. The buyer shortfall is $8.5T.

See the demand calculation
Sources and exact values

Source files · layers_v4.csv · rows_v4.csv · own_blocks_v4.csv

stage1-v3/requirements/layers_v3.csv; stage1-v3/repair/baseline/layers.csv (today)

Copy this prompt

Copy the selected text using your keyboard or touch menu, then paste it into your assistant.