Section B Supporting Statement Final with all clearance initials.wpd

SUPPORTING STATEMENT

U.S. Department of Commerce

U.S. Census Bureau

Government Finance Forms

OMB Control No. 0607-0585

Part B. Collection of Information Employing Statistical Methods

1.Universe and Respondent Selection

For our annual survey, we will use a sample of approximately 11,300 non-school local governments selected with probability proportional to size in order to estimate data for governments other than school districts¹. For 2008, the sample distribution for non-school local governments will be about:

Table 3. Sample and Universe Counts of Non-Education Local Governments
Type of Government	Sample (For Fiscal Year 2008 Survey)	Universe
Counties	1,736	3,034
Municipalities	2,904	19,429
Townships	2,125	16,504
Special Districts	4,579	35,052
Total	11,344	74,019

The sample of 11,344 non-school governments for the 2008, 2009, and 2010 Fiscal Year surveys are a state-by-type of government sample derived from the universe of the 2002 Census of Governments. The sample for 2009 - 10 may be smaller as we consider other methods for sampling townships, special districts, and small municipalities. For our public employee-retirement systems, Forms F-11, and F-12, we will use a sample of approximately 1,077 state and local retirement systems selected with probability proportional to size in order to produce state area estimates at the state and local level.

Table 4. Sample and Universe Counts of Public Employee-Retirement Systems
Level of Government	Sample for Fiscal Year 2008 Survey	Universe
State Systems	221	221
Local Systems	856	2433

The response rates for the 2006 Annual Finance Survey, our most recent completed survey, were:

1. for county governments; 87%

2. for cities and township governments; and 86%

3. for special district governments. 71%

The overall response rate for the 2006 Annual Finance Survey, Retirement Phase, was 93.4%. The weighted response rate was 99.2%.

2. Procedures for Collecting Information

Sampling

The sample for the 2008 Annual Finance Survey will be the same sample as was used for 2006 with the additional birth units added. The annual surveys for the 2009 and 2010 surveys will use a state-by-type of government sample drawn from the 2007 Census of Governments. The initial certainty criteria are as follows:

1. All county governments with a 2007 population of 100,000 or more;

2. All cities with a 2007 population of 75,000 or more;

3. All townships in New England and Middle Atlantic states with a 2007 population of 50,000 or more;

4. All special districts that satisfy any of the following conditions:

5. It has a 2007 full-time equivalent employment of 1000 or more;

6. It has a function code designating it as providing mortgage credit, electric power, gas supply, or public mass transit services.

7. It has either a minimum of $20 million dollar long-term debt outstanding or a minimum of $10 million dollar in total revenue or total expenditure; or

8. It is the sole special district providing that service within a state.

Non-certainty units are selected for the sample based on a probability proportional to its size, as determined by long-term debt and total expenditure of the unit. Research into methods for reducing the sample size are being conducted.

The annual state and local public employee retirement systems for 2008, 2009 and 2010 will use a state area by level (state-local) sample drawn from the 2007 Census of Governments. The initial certainty criteria are as follows:

1) All state government retirement systems;

2) Systems with total holdings and investments of $50 million or more;

3) Retirement systems of Finance Jacket Units (very large general purpose governments).

4) Systems in a state that had fewer than 10 local retirement systems with a probability not equal to one

5) Systems with total earnings on investments greater than $8 million, or total benefits paid and withdrawals greater than $15 million.

Non-certainty units in the annual state and local public employee retirement systems phase are selected for the sample based on a probability proportional to its size.

Estimation Procedure

The Annual Finance Survey calculations employ a regression estimator for all estimates except long-term debt and capital outlay items. Long-term debt and capital outlay data use a simple unbiased estimator. Retirement Survey estimates also use a simple unbiased estimator.

Degree of Accuracy

State Area Estimates of Local Government Totals. For state area estimates of only local government totals, the finance sample for Annual Finance Surveys is designed to produce totals with a relative standard error of three percent or less or five percent or less on the following 11 major financial items:

Three percent or less:

• Long-term debt outstanding

• Total revenue

• Total expenditure

• Criminal justice expenditure

Five percent or less:

• Education expenditure

• Highways expenditure

• Health expenditure

• Housing expenditure

• Total capital outlay expenditure

• Utilities expenditure

• Welfare expenditure

State Area Estimates of State and Local Government Totals. With the addition of state government data to the local government information, the finance sample for the Annual Finance Survey generally yields state area estimates with a relative standard error of one percent or less on total revenue and total expenditures and under two percent for long-term debt outstanding.

Unusual Problems

None.

Use of Periodic Data Collection Cycles

These data are critical components for calculating the National Income and Product Accounts, which require a minimum of annual data.

3. Methods to Maximize Response

We use several methods and resources to maximize response in the collection of government finance data, including: follow-up mailings; telephoning nonrespondents; central data collection techniques; Internet data collection; and reports obtained from the Census Bureau's Federal Single Audit Clearinghouse (SAC). We are trying to improve access to the materials in the Federal SAC. Because it would help minimize respondent burden and maximize response in this survey. The rules governing release of data in the Federal SAC limit its usefulness for this survey.

4. Tests of Procedures or Methods

The Census Bureau conducted formal testing for its local government finance program in 2004 and implemented the results in subsequent survey years. As a general practice we make improvements based on consultations and respondent feedback.

5. Contacts for Statistical Aspects and Data Collection

Carma R. Hogue (301-763-4882) of the Office of Statistical Methods and Research for Economic Programs monitored the statistical aspects of the sample design. Jennifer Whitaker (301-763-2823), Chief, Public Finance Analysis Branch-A, Governments Division, is available to answer any data collection questions.

List of Attachments

(1) Forms F-5, F-11, F- 11(S), F-12, F-12(S), F-13, F-25, F-28, F-29, F-32, F-42

(2) Cover Letters

1Elementary and secondary school system finance data are collected under authority of a separate submission; see OMB clearance 0607-0700, expiration date 05/31/2009. The elementary and secondary education finance data collection is a joint undertaking of the National Center for Education Statistics and the Census Bureau. The survey coverage for this annual activity is the universe of elementary and secondary education systems. This coverage allows both agencies to fulfill their missions. For years in which we conduct a Census of Governments, such as 2007, we use the universe data. For years in which we are conducting a sample survey, such as 2008, 2009, and 2010, we use the universe for independent school systems. For dependent school systems – those that are not a separate government, but which we classify as part of a general purpose government – we use the government units chosen in the non-school local government sample.