Two decades of demand in a single index.

The Concrete Index

For each month since 2004, the index combines trailing-twelve-month value and trailing-twelve-month count of new construction awards (renewals excluded by selection method; awards of $500M and above excluded so single megaprojects don't move the cycle reading). Award notices are first deduplicated to one record per procurement (the City Record pin, keeping the largest published amount), the same rule the “Awarded” page uses, so the same award is never counted twice. The two components are expressed relative to their 2019 baselines and blended geometrically; 2019 = 100. Each baseline is the mean of the twelve trailing-twelve-month readings dated in 2019, a mean of overlapping year-long windows reaching back to February 2018, rather than the calendar-year 2019 sum. The two bases differ by a few percent, so the level of the series depends on which one is used; this page uses the trailing-twelve base throughout, including in the cost-adjusted variant. Dollars are nominal; the index reads demand activity rather than inflation-adjusted volume.

Why this differs from the “Awarded” page total. The trailing-twelve award figure on the “Awarded” page sums every construction award in the window after the per-procurement dedupe, including renewals and awards of $500M and above. This index deliberately strips renewals and megaprojects out to isolate the cycle in genuinely new work. Both numbers are computed on the same deduplicated universe and are correct for their purpose, so the two pages will show different trailing-twelve totals by design.

Cost adjustment

The nominal index counts dollars at face value, so it cannot tell rising prices from rising volume. The adjusted series deflates it. Two Bureau of Labor Statistics price series carry the cost signal: materials, the national producer price index for inputs to construction industries, goods (WPUSI012011); and labor, New York State construction average hourly earnings of production and nonsupervisory employees, not seasonally adjusted (SMU36000002000000008). Materials stays national because construction inputs trade in national markets, while labor is localized to New York because construction wages are set regionally. The labor series uses the production-and-nonsupervisory definition because it reaches back to 2004, matching the award record, and is the methodological twin of the national series it replaced. Localizing the labor leg ties the deflator to New York wage growth rather than the national average. Where the two diverge, the cost-adjusted index reads above the all-national blend when New York wages have run slower since 2019 and below it when they have run faster.

Each series is rebased to its own 2019 twelve-month average = 100, then blended into one monthly deflator weighted 60% materials and 40% labor, a split that reflects the materials-heavy mix of the City's general construction work while still carrying the wage signal. Because both series are monthly, the deflator needs no interpolation; the most recent one or two months are forward-filled at the last published value until BLS posts them, and those points are flagged in the data. Every award month's new-award value is divided by that same month's deflator before the trailing-twelve sum, so a 2021 award and a 2025 award are each discounted by the cost level that actually prevailed when the City committed the money, not by a single blanket factor. The cost-adjusted index then applies the same geometric blend with award count, holding the count component unchanged. By construction the 2019 real total equals the 2019 nominal total, which the page uses as an internal check.

Caveats. The labor leg is measured for New York State, but the materials leg is national, so the deflator still approximates local materials cost rather than measuring it. Award value is a floor on real activity, since the published amount is the award notice, not the final registered contract. The 60/40 weighting is a deliberate approximation of input mix, not a measured cost build-up. The nominal index remains the headline reading; the adjusted series is a separate, disclosed view, and its latest months inherit the forward-filled cost values noted above; New York labor publishes one month behind the national materials figure, so the most recent month is forward-filled.

Seasonality

For each complete calendar year, each month's share of that year's new-award value is computed; the profile shows the median share across years, scaled so 1.0 equals an average month. Medians keep single unusual years from distorting the profile.

Pipeline conversion

Projects are matched across consecutive capital plan snapshots (roughly four months apart) by agency and FMS ID, pooled over the two most recent transitions. A project "advances" when its delivery phase moves forward in the sequence pending → pre-design → design → procurement → construction → close-out. Projects that leave the dataset between snapshots are not counted.

Budget escalation

Projects present in both the earliest available plan snapshot (May 2023) and the most recent one, with budgets above $1M in both, are matched by agency and FMS ID. Escalation is each project's total-budget growth across that span; the headline figure is the median. Budget growth blends scope changes with cost escalation. The public record doesn't separate the two, so neither does this page.

Dated payment timing

Two sections read a separate, same-origin dataset, /data/contract-search.json (the file the “Contracts” page uses), falling back to a bundled sample if the live file is absent. It holds registered construction contracts, each carrying its scope description and the City's procurement identification number, both searchable on that page; only those with confirmed Checkbook NYC disbursements carry dated payment timing, and only those are used here. The payment curve averages percent of contract value paid at each whole month since registration across the dated set, with the three agencies holding the most dated contracts overlaid; the contract set behind each month changes as contracts enter and finish, so the line is a population average rather than one contract's path, the horizon is capped at 60 months, and dollars are nominal. The schedule-pace scatter plots each agency's mean schedule-elapsed against its mean paid-to-date; schedule elapsed is start-to-today over start-to-end, clamped to 0 to 150 percent, and the end date is the registered or planned end, not actual completion, so the read is directional rather than a measure of completion. Both sections require at least eight dated contracts per agency before an agency is shown.

Sources

Award history: City Record Recent Contract Awards (qyyg-4tf5), 2003→present. Capital plan: Capital Projects Dashboard (fb86-vt7u), snapshots May 2023→present. Both on NYC Open Data, queried live on every page load. Construction cost: BLS series WPUSI012011 (national materials) and SMU36000002000000008 (New York State construction labor, NSA), 2004→present, fetched server-side by the nightly archive and read here from a same-origin snapshot.