CommentairesÉclairagesPerspectivesDécryptages
Archives
Correspondant à ces filtres : 32 articles
Articles
- PerspectiveMacroeconomics & Markets
The measurement has to exist before the policy, because no earlier period can be collected afterwards
What a statistic can answer is decided by its base, its frequency, its territory and the date on its newest observation, and all four are fixed before any policy question arrives. The complication is that a record built this way reads as routine paperwork until something needs an answer it never arranged to have.
- PerspectiveInfrastructure
The boundary of a public investment appraisal is drawn where the measurement stops
The capital cost in a public investment appraisal has a published national counterpart, nonresidential construction spending at 1309 billion dollars a year; the benefit stream it is weighed against has none. The discount rate decides the rest: at 4.29 percent a dollar arriving in year thirty is worth 28.4 cents, and at the 0.89 percent of 2020 it was worth 76.7 cents.
- PerspectiveIndustrial Policy
Productive capacity names a quantity that no series on this site reports
Of the fifty-eight annual series in the file behind this site, one carries the word capacity in its title, and it reports a percentage of capacity rather than an amount of it. What stands in are four measures of four different things, and they moved in different directions over the same three years.
- PerspectiveTrade & Investment
Foreign delegations come to Washington to meet a counterparty that is not there
Twenty of the fifty-one jurisdictions in the federal collections file levy all five of the taxes it names, and two levy only two of them. The body that decides whether a foreign project works is usually a state, and it is not in the room.
- PerspectiveWorkforce & Skills
The institutions that match workers to work appear in no series that measures the match
Job openings averaged 11.2 million a month in 2022 and 7.1 million a month in 2025, and the quits rate ended the window at 2 percent of employment, under its 2020 reading. Prime-age participation finished at 83.6 percent, its highest, so whatever is weakening is not weakening everywhere the statistics look.
- CommentaireIndustrial Policy
Announced investment is not a statistic and industrial policy should stop being scored on it
Construction spending on manufacturing plant rose from an annual rate of 82 billion dollars in 2021 to 215 billion in 2025, while the manufacturing output index moved from 97.6 to 97.3 on a 2017 base of 100. Lag complicates the comparison: a plant paid for in 2024 reaches an output index years later, which argues for waiting rather than scoring the policy on its announcements.
- PerspectiveIndustrial Policy
Public purchasing is an industrial instrument, and none of these series counts it as one
A government's order book transfers demand risk in a way a grant cannot, which is the reason it would work as industrial policy. What complicates the case is that the spending which is counted rose steeply, plant construction going from 75 billion dollars in 2020 to 244 in 2024, and manufacturing output is still below its 2017 base.
- PerspectiveGovernance & Institutions
A state can only set a target on something its own accounts already count
A target can be written only on a quantity somebody already publishes, so the statistical system settles in advance which problems a government is able to own. The complication is that absence is not neutral: where no series covers a subject, the nearest published count is promoted into the gap, and the policy ends up governing whatever that count measures.
- PerspectiveInfrastructure
An infrastructure delivery model allocates risk, capital and blame, and it reduces none of the three
A delivery model is a contract about where the risk, the capital and the accountability sit, and only two of those can be priced and sold. The ten-year yield moved 3.4 points in five years while the record of what was built stayed one undifferentiated line, so the comparison between models is one the public record cannot check.
- PerspectiveMacroeconomics & Markets
Institutional memory in economic policy is held in definitions, not in people
What an institution loses first is not an event but a definition: the unit a figure is stated in, the base year of an index, the rate regime a career was formed inside. Memory is not free either, because an assumption carried too long is its own failure, and only a written record separates the two.
- PerspectiveDevelopment Finance
The development finance architecture measures itself in the lender's currency and the lender's year
Concessional lending is judged against a market rate, and market rates moved: the ten-year Treasury averaged 0.89 percent in 2020 and 4.29 in 2025, and the broad dollar index sits 9.6 points above its lowest reading, in 2021. The complication is that an institution unable to report a figure on signing day cannot be funded, which makes the mismatch a constraint rather than a failure of will.
- PerspectiveIndustrial Policy
Regulation can change what an asset does; only ownership decides that it exists
Ownership gives a state two rights no rule confers: the decision that an asset exists, and the claim on whatever is left over once it does. What complicates the case is price, because the ten-year Treasury rate averaged 0.89 percent in 2020 and 4.29 percent in 2025, and a state that owns must raise the money before it decides anything.
- PerspectiveIndustrial Policy
Institutional quality is several separable capacities, and the score that ranks them is mostly a choice of weights
Operationally the term names separable things: a task specified, a cost stated, a time stated, and a published record of whether the work was done. A composite removes the units before it adds them, and on this site's own state index the weightings a person could defend move one jurisdiction across thirty places.
- PerspectiveIndustrial Policy
Industrial strategy is a decade-long commitment made by institutions that are not built to hold one
Spending on manufacturing plant rose from a yearly rate of 75 billion dollars in 2020 to 244 in 2024, lifting the factory share of all nonresidential building from 8.8 percent to 18.8. Manufacturing employment reached its high for these six years in 2023 and is lower now, which is what it looks like when a country can order buildings faster than the institutions to run them.
- PerspectiveGovernance & Institutions
Reform of a multilateral institution changes behaviour only when it changes what the institution must count
Reform is argued as a question of votes and chairs, while behaviour is set by the balance sheet, the appropriation and the measure that decides eligibility. The statistics that would test any of it are national, and this site holds no comparable published series for the institutions themselves.
- CommentaireHousing Supply
Federal policy cannot raise housing supply and should pay for authorisations instead
Idaho authorised 0.98 new private housing units per thousand residents in July 2026 and California 0.21, a ratio of 4.66, under the same federal tax code and the same policy rate. Idaho's house-price index has still multiplied 8.78 times since 1980 against California's 9.75, so authorising more has not by itself held appreciation down.
- PerspectiveEconomic Security
A supply chain is governed on private books, and the public statistics reach only its edges
American business held an average of 2,663 billion dollars of inventory across 2025, 35.1 percent above 2020, and that buffer is the working instrument of resilience. The series records its size and nothing else: not who holds it, not against which supplier, not where the next tier sits.
- CommentaireWorkforce & Skills
Administrative capacity should be counted in decisions per qualified head, not in the occupational group that carries its name
Office and Administrative Support is the largest occupational group in the country, 17,753,410 jobs of a 51-state total of 155,495,750. The District of Columbia, where administration is the industry, ranks last of the 51 on it, at 76.871 such jobs per thousand against Tennessee's 129.94.
- CommentaireEnvironmental Permitting
Whether permitting is slow can only be answered by counting the applications still waiting
Housing units authorised by permit fell from an annual rate of 1734 thousand units in 2021 to 1431 in 2025, a drop of 303. That series counts decisions leaving the permitting system and never the applications sitting inside it, and a count of departures is consistent with any waiting time at all.
- PerspectiveRegional Development
Read a state government the way you would read a credit, not a brochure
State tax collections per resident run from 2,279 dollars in Alaska to 8,028 in Vermont, and the spread describes which layer of government collects rather than how heavily anyone is taxed. The useful questions are what the base is, what it is exposed to, and what the state does not levy at all.