Data Sources

NYS Comptroller (OSC)

The NYS Office of the State Comptroller collects Annual Financial Reports (AFRs) from local governments across New York State. NY Benchmark imports this data for all 62 cities. Of these, 61 have filed at least once since 1995. The remaining city — New York City — has its own Comptroller and is not part of the OSC reporting system. We plan to import NYC data from Checkbook NYC in a future release.

As of the most recent filing year (2024), 57 cities have submitted data. Four cities are late or non-filing. The OSC dataset includes revenue, expenditure, and balance sheet data at the fund and account-code level, totaling over 647,000 individual observations.

US Census Bureau (ACS)

Population, income, poverty, and housing data comes from the American Community Survey (5-year estimates) via the Census Bureau API. This covers all 62 cities (including NYC) since 2012, totaling over 14,000 observations.

ACS 5-year estimates are based on survey samples and carry margins of error. For small cities, these margins can be significant. We use point estimates throughout the application.

How Metrics Are Calculated

The three headline metrics used in city rankings and entity dashboards are derived from OSC and Census data:

Fund Balance %
Unassigned Fund Balance (account code A917) divided by Total Expenditures, expressed as a percentage. A higher fund balance generally indicates stronger fiscal reserves.
Debt Service %
Total Debt Service (all accounts in the Debt Service category) divided by Total Expenditures, expressed as a percentage. A lower percentage typically indicates less debt burden.
Per-Capita Spending
Total Expenditures divided by population (from Census ACS data). Provides a normalized comparison of spending across cities of different sizes.

All-Fund Approach

Expenditure and revenue totals include all fund types (General, Highway, Water, Sewer, Enterprise, etc.) with two exclusions:

  • Trust & Custodial Fund (T-fund) — Cities in Westchester and Nassau counties act as tax collectors for county and school district taxes. These pass-through amounts are reported in the T-fund and can inflate apparent expenditures by 40-50%. We exclude the entire T-fund to remove these pass-throughs.
  • Interfund Transfers — When cities move money between funds, OSC reports these as expenditures ("Other Uses") and revenues ("Other Sources") at the fund level. Cities' own ACFRs eliminate these in government-wide statements. We do the same to prevent double-counting.

Fund Structure Variation

Cities organize their finances differently. Some run water and sewer operations through the General Fund (A), while others use dedicated Enterprise (E), Water (F), and Sewer (G) funds. Debt service may appear in any fund. The all-fund approach handles this correctly by including all legitimate spending regardless of fund structure.

Known Limitations

  • Late and non-filing cities — Not all cities file their Annual Financial Reports on time. Four cities are currently late, with the most significant case being Mount Vernon, which lost its credit rating due to years of non-filing. View filing compliance details.
  • NYC not yet included in OSC data — New York City operates under its own Comptroller and does not file with the OSC. NYC data will be imported from Checkbook NYC in a future release.
  • GASB 54 and fund balance classification — The "Unassigned Fund Balance" metric (account code A917) only exists from 2011 onward, when GASB Statement No. 54 reclassified fund balances from Reserved/Unreserved into Nonspendable, Restricted, Committed, Assigned, and Unassigned categories. Some cities also reclassify fund balance between categories year to year. For example, Buffalo reported no Unassigned Fund Balance for 2018-2020, instead categorizing its entire General Fund balance as Assigned or Restricted during those years.
  • Early year coverage — Data coverage is sparser in the earliest years (1995-2000) as fewer cities had filed electronically.
  • Census margins of error — ACS 5-year estimates are survey-based and carry margins of error, particularly for smaller cities.

Open Source

NY Benchmark is fully open source under the MIT license. The source code, data import scripts, and issue tracker are all available on GitHub. Contributions, bug reports, and suggestions are welcome.