Open data on public expenditure — datasets, categories and a SPARQL endpoint for research and analysis

Understanding US Research Spending Through the NSF

US federal spending on scientific research reaches universities, laboratories, companies, schools and public agencies through many channels. The National Science Foundation is one of the most important, supporting basic research, engineering, education and the infrastructure needed to produce new knowledge. Its grants can fund a single investigator, a national facility, a multi-university collaboration or a major technology programme.

For researchers, journalists and citizens, the difficult part is rarely finding the NSF’s headline budget. The harder task is tracing how an appropriation becomes an award, identifying who receives the money, distinguishing commitments from actual payments and comparing one year with another. Open spending data makes those steps more practical.

Publicspending.net is designed for this kind of investigation. Its standardised datasets, visualisations, bulk downloads and semantic tools can help users examine public payments across jurisdictions. An Australian reader can use the same approach to compare American research funding with the Australian Research Council, the National Health and Medical Research Council and Commonwealth grant programmes.

What The National Science Foundation Supports

The NSF funds research across the physical sciences, biological sciences, mathematics, computer science, engineering, social science and education. Its directorates support areas such as advanced materials, climate science, robotics, cybersecurity, astronomy, oceanography and biotechnology. Much of this work is exploratory: the commercial or social result may be years away, and a successful project may produce knowledge rather than a saleable product.

The agency also pays for shared research infrastructure. This includes scientific instruments, supercomputing capacity, observatories, laboratories, data resources and large facilities used by researchers from multiple institutions. In this respect, an NSF award may strengthen an entire field rather than benefit only the named university or principal investigator.

NSF spending also reaches education and workforce development. Programmes can support postgraduate training, science teaching, fellowships, public engagement and pathways into technology careers. The agency’s role therefore extends beyond laboratory research. It helps shape the supply of researchers and technical workers who later contribute to universities, government, health services and private industry.

Budget Authority Is Not The Same As Cash Paid

A federal research budget usually begins with congressional appropriations. That legal authority allows an agency to commit money for approved purposes, subject to the conditions set by Congress. The NSF then makes awards, enters contracts and pays its operating costs. These transactions appear in federal spending systems under different financial categories.

An obligation is a legally binding promise to spend. For a multi-year grant, the full award may be obligated even though the recipient draws the money gradually. Outlays, or payments, show when funds actually leave the government. A search that treats obligations as cash expenditure can therefore make a research programme look larger in one year than the payments recorded for that year.

The distinction matters when comparing the United States with Australian public finance. Australian Commonwealth reporting also separates appropriations, commitments and actual expenditure, with agencies operating under the Public Governance, Performance and Accountability Act 2013. A sound comparison should use equivalent measures, such as annual payments against annual payments or award commitments against award commitments.

The NSF’s annual budget can also be affected by continuing resolutions, carry-forward funds, special legislation and changes in accounting treatment. The CHIPS and Science Act authorised substantial future investment in American science and technology, but authorisation is not identical to the amount ultimately appropriated and spent. Data users should check the fiscal year and transaction type before drawing a trend line.

Where NSF Research Dollars Go

Universities receive a large share of NSF research funding because they host investigators, laboratories and graduate programmes. Awards may be made to a lead institution while allocating money to partner universities, research centres or specialist contractors. A university’s apparent total can therefore include funds passed to collaborators, while another institution may appear separately in the same project’s records.

Private companies, nonprofit organisations, state and local entities and federally funded research and development centres can also receive NSF money. Small technology firms may participate through programmes intended to move promising research towards commercialisation. These awards are different from conventional academic grants and should be analysed separately if the goal is to measure university research.

The location attached to an award deserves careful attention. It may represent the recipient’s headquarters, the performing organisation, a payment address or an administrative office. A grant recorded in Washington, DC, for example, does not necessarily describe research physically carried out there. The same caution applies to awards recorded in California, Massachusetts or Illinois.

For Australians, this is similar to the difference between the registered address of a university and the location where a project operates. A grant administered by a campus in Sydney might support fieldwork in regional New South Wales, while a Melbourne-based company may contract suppliers interstate. Administrative geography is useful, but it is not a perfect map of economic or scientific activity.

Reading Recipients, Awards And Suppliers

An NSF award record normally contains identifying information about the recipient, the award amount, the programme, the start and end dates and the relevant funding office. Depending on the source, it may also include the award title, abstract, principal investigator, congressional district, assistance listing or transaction history. These fields allow users to move from a broad agency total to a specific research portfolio.

Recipient names are often inconsistent across years. A university may appear under a legal corporate name, a shortened trading name or a separate foundation. Campuses within one system can be treated as individual recipients or grouped under a parent organisation. Standardisation is essential before ranking institutions or calculating growth rates.

Publicspending.net’s value lies in bringing records into a common structure and making them easier to query. A researcher can use tables and graphs for an initial overview, download records for statistical work or use the SPARQL endpoint to explore relationships between agencies, organisations, places and spending categories. Ontology-based data can be especially useful when the same concept appears under different labels in separate sources.

Every result should still be checked against the original federal record. A portal can improve discovery and comparison, but it cannot remove the need to inspect definitions, update dates, missing fields and duplicate transactions. Reproducible analysis should preserve the query, source version, fiscal-year selection and filters used to create each figure.

Questions That Produce Better Results

A useful investigation starts with a specific measurement question. “How much did the NSF spend?” could mean the agency’s total outlays, the value of new awards, the money committed to a particular state, or payments to a single university. Each interpretation requires different filters and can produce a different answer.

Researchers should decide whether to include all NSF directorates or focus on a subject area. A computer science analysis may examine the Directorate for Computer and Information Science and Engineering, while a climate study might combine awards from geosciences, biological sciences, engineering and other programmes. Programme boundaries can change, so historical comparisons need documentation.

Time is another important filter. Federal fiscal years run from 1 October to 30 September, unlike the Australian financial year, which runs from 1 July to 30 June. A US fiscal-year 2024 result does not align neatly with Australia’s 2023–24 reporting period. Converting dates without explaining the difference can create misleading comparisons.

It is also useful to separate competitive research grants from contracts, fellowships, facility operations and administrative spending. A large infrastructure award may dominate a year’s total without indicating a broad increase in investigator-led research. Breaking totals into award type, directorate and recipient category makes the underlying pattern easier to interpret.

Comparing American And Australian Research Systems

The United States has a large and decentralised research system, with the NSF operating alongside agencies such as the National Institutes of Health, NASA, the Department of Energy and the Department of Defense. Australia’s funding landscape is smaller and more concentrated. The Australian Research Council supports non-medical research, while the NHMRC focuses on health and medical research, and government departments administer targeted programmes.

The difference can be seen in university markets in Sydney, Melbourne, Brisbane, Canberra, Perth and Adelaide. Australian institutions often compete for national grants while maintaining international partnerships. An Australian researcher examining NSF data may therefore be interested in collaborations, co-authorship networks and whether US awards support projects involving Australian universities or facilities.

Local legislation and reporting practices shape what is visible. Australian Commonwealth procurement is commonly searched through AusTender, while grant information may be published through departmental portals or the federal grants system. The PGPA framework sets expectations for accountable authority, performance and reporting, but public data does not always provide the same granularity as a detailed transaction file.

Everyday economic conditions also affect how research funding is experienced. A grant that supports a laboratory in Boston or San Francisco pays wages, rent, equipment suppliers and specialist services in markets with very different costs from Canberra or Hobart. Australian readers should avoid treating a US dollar figure as a direct measure of research capacity without considering exchange rates, salaries, property costs and institutional overheads.

Turning Spending Records Into Public Insight

Spending data can reveal concentration, continuity and change. A sequence of awards may show that a university has developed strength in quantum computing, marine science or artificial intelligence. A cluster of suppliers may indicate a regional technology market. A sharp fall may reflect the end of a facility project rather than a reduction in scientific quality.

The data can also support accountability. Users can examine whether promised funding reached the intended recipient, whether an award was modified, how long a project remained active and whether the distribution of funds matches stated policy priorities. Journalists may combine award records with university disclosures, company filings, patent data or published research to investigate outcomes without assuming that expenditure alone proves success.

Visualisations are helpful when they expose structure rather than decorate a total. A map can show recipient locations, a timeline can display obligations and payments, and a network graph can connect agencies with institutions and suppliers. Each visual should identify its unit, date range and source, particularly when users may compare it with Australian government figures.

The most responsible interpretation recognises the limits of administrative data. A database can show who received an award and how much was recorded, but it may not capture unpaid collaboration, scientific failure that still produced valuable knowledge, indirect economic effects or benefits shared across countries. Public spending records are a foundation for inquiry, not a complete scorecard of research value.

Use Publicspending.net to trace NSF awards from agency totals to recipients, places and programme categories. Download the underlying records, document your filters and compare equivalent measures with Australian research funding data. Open, carefully interpreted evidence gives citizens and researchers a stronger basis for understanding how public money builds scientific capacity.