City Of Los Angeles Geographic Revenue Analysis For Fiscal Year 1998-1999

Note this file was converted to html from a word document provided through the Harbor Study Foundation from LA City 12-18-00 and Indexed

Preliminary data was released December 2000 and the final report in February 2001. The changes between the Draft and the final version as marked green strikeout for deletions and red for insertions. Grammatical and layout changes are ignored

TABLE OF CONTENTS

1 Introduction

2 Methodology

Data Sources
Geocoding
Population Estimates

3 General Fund Revenues

4 Special Purpose Revenues

5 Additionally Budgeted Funds

6 References

LIST OF TABLES

City of Los Angeles Population

LIST OF MAPS

Geographic Study Areas The report failed to include a map, we have linked the map from the petition.

2 SECTION 2

METHODOLOGY

Approximately 80 revenue items in the operating budget and several additional budgeted revenues are evaluated in this report. Revenue characteristics are described, a methodology for evaluation is provided, analysis is conducted and additional issues are identified for each revenue source.

Several methodologies to distribute revenues geographically are employed throughout this analysis. Methodologies were developed based on the characteristics of the revenue fund, including revenue source and jurisdictional requirements. The two most common methodologies used in this study are geocoding and source based on population, as described below in detail. Other methodologies are described in detail within the appropriate sections of this report as they apply, including generation based on property characteristics and proxies when original source data are not available. To index

DATA SOURCES

Where possible, data were drawn from the original source. Various City departments have provided program-level data used to manage permit, fee, and tax payments. For revenues such as the Transient Occupancy Tax and the Business Tax, address and the amount of revenue generated at each address were extracted for every account within the reference year of FY 1998- 99. For other data sources, primary data are not available. Cable Franchise fee data, for example, are not available for each household . In these cases, program area data were used, as these are the most basic data available. In some cases, proxy data were used because neither original data nor summary data with a geographic identifier were available. Telephone companies, for example, provide no data other than the total amount of tax due to the City. Since this data is useless in a sub-regional analysis and account level data are not available, a proxy was developed to estimate the amount of specific types of revenues that are generated in certain areas of the City. Finally, several state and federal grant programs assign revenues based on demographic data. Appropriate demographic data were used in these cases to assign revenues.

In all cases, data from program management databases do not match the actual revenues reported in the City of Los Angeles Budget for the Fiscal Year 2000-2001. Actual data figures for FY 1998-99, the most recent year for which actual data are available, are reported in the 2000- 2001 budget. These figures are provided by the City Controller and represent corrections and revisions. Raw data collected from the various City departments are close to the reported Actual data, but do not precisely match. Since data in the FMIS system do not contain geographic identifiers, raw data were the next most appropriate source of data for this analysis. To index

GEOCODING

Geocoding is a process that assigns geographic information to a dataset. The dataset must initially contain some type of information that will allow for assignment of geographic information. Various forms of geographic information can be assigned in a geocoding process, ranging from very precise coordinates like an x/y coordinate to broad regional designations such as a census tract. Depending on the quality of source data, the initial type of geographic information in the dataset, and the media on which the data is recorded, various geocoding methods can be used to assign additional geographic information.

Address Matching

Address matching is a useful form of geocoding. The parts of an address, such as street number, street name, and street type, can be compared to a map to determine a specific location with x/y coordinates. This produces an accurate location for the address in relation to other streets and, at a broader level, to other geographic areas. On paper, this is referred to as a pin map. This can also be accomplished on a computer using software called a geographic information system (GIS). The software can process address data, using the various parts of an address to automatically determine a location and assign a latitude and longitude coordinate to the address. This coordinate information can then be used to determine whether that address is located within other geographic areas, such as a city or a police district or a flood zone.

Much of the data obtained for this revenue analysis were address matched. Various City departments provided databases that contained addresses associated with payment of a particular fee or tax. Addresses were then processed using ArcView GIS software. Two address matching processes were used in this study, one involving the Thomas Bros. street file and the other using Code 1 software in conjunction with the Thomas Bros street file.

Thomas Bros. Street File

The GIS compares address data in a database to a street graphic file. Many different street files are available for use, including products prepared by the U.S. Census Bureau. In Southern California, the Thomas Bros. street map is a standard and comprehensive street file that is very useful in address matching projects. The Thomas Bros. street file was selected for use in this project due to the broad range of City geographic data that can be associated with the street file, including Police Department bureau boundaries, cable service areas, and census tracts.

For this study, the Thomas Bros. street file was imported to the ArcView GIS software. Addresses from each database containing address information were then compared against the Thomas Bros. street file in ArcView to produce an x/y coordinate for each address. Addresses were compared at the most strict level of analysis, requiring that street numbers, street names, street numbers, and zip code in the database achieve a high degree of accuracy in their match to the street file.

Any address that did not match due to poor address information was visually reviewed. Addresses do not match for many reasons, including spelling errors, the addition of new streets that have not appeared in the street file, and address locations that are not actual streets. For example, many berths in the Los Angeles World Port are acceptable mailing addresses that cannot be address matched to a GIS street file. Addresses that included a berth in the Harbor were automatically assigned to the Harbor study area. Minor flaws, such as street name spelling errors, were corrected and the unmatched addresses were rematched. Additional research was conducted for other unmatched addresses to determine the reason and to provide for corrections. Where no solution is available, addresses remained unmatched.

Code 1 and Zip Plus Four

One address dataset was processed using a different address matching methodology. The Department of Water and Power (DWP) provided a database that contained more than 1.5 million records containing the address and utility information for every account in the City of Los Angeles. Due to the size of this database, the Information Technology Agency (ITA) address matched the data using a computer than could process such a large dataset. Before address matching the dataset against the Thomas Bros. street file as described above, ITA first processed all 1.5 million records through a software package called Code 1.

Code 1 is an address correction software certified by the U.S. Postal Service that evaluates address information in a database, then produces a compete address with a zip plus four zip code that is an accurate postal address. ITA used Code 1 to ensure that most of the addresses in the DWP database would match to the Thomas Bros. street file. After processing the DWP database through Code 1, ITA proceeded to address match the database against the Thomas Bros. street file as described above.

Geographic Regions or Areas

Once an x/y coordinate has been assigned to an address, a range of additional geographic areas can be assigned to that address. GIS is used to compare the x/y coordinates in a database to a geographic area to determine whether that coordinate falls within the area. If the coordinate falls within the area, the name of the area and other information about the area are attached to the coordinate. For example, a census tract or police bureau could be assigned to a record once an x/y coordinate has been assigned. The GIS will compare every record in a data file against each selected geographic region to assign the appropriate area to that record.

For the purpose of this study, a geographic file was prepared that shows the regions of the three areas that would result in the event that the San Fernando Valley and Harbor areas secede from the City of Los Angeles. The San Fernando Valley and Harbor geographic areas match the proposed secession areas as proposed to LAFCO. Once the address-level data were address matched, the x/y coordinates assigned by the GIS were then used to determine which of the three study areas that address was located in. To index

POPULATION ESTIMATES

Several revenue sources are distributed based on City population estimates. As a result, any methodology to analyze these revenues requires population data. The 1990 Census is the last official population count conducted by the federal government. Most estimates or projections of population for subsequent years are based on these data. The 2000 Census is nearly concluded and preliminary data will arrive in December 2000. Since this data will not be available until after the completion of this study and since year 2000 population figures would not match this study’s reference year of 1998-1999, alternate population data sources were identified.

Several revenues received by the City are distributed by the State of California based on population. The State Department of Finance (DOF) prepares annual population projections that are used by various state agencies to distribute these revenues. Data are reported for various cities and counties, but DOF does not provide data at the census tract level. Since the nature of this study is to evaluate revenue generation regionally within the City, DOF data are of very limited use.

The City of Los Angeles Department of City Planning (DCP) prepares annual population estimates for various City uses. DCP prepares its population estimates using DOF data for the City and estimating the distribution of that population across the City’s census tracts using school enrollment data, building permit data, and other population indicators. The DCP population estimates at the census tract level, then, allow for the aggregation of population data within smaller regions of the City. Using DCP data, Table 2 shows that the estimated population within the three study areas. To index

To index
To Main Index