# Report on Summer 1993 Surveys Used To Determine Cost-of-Living Allowances in Selected Nonforeign Areas

> Briefs, arguments, decisions, and more.

URL: https://www.frixlaw.com/law-library/documents/fr%3A94-12463

## Record

- **Collection:** Federal Register
- **Document type:** Uncategorized Document
- **Published:** May 26, 1994

## Text

SUMMARY: This notice publishes the ``Report to OPM on Living Costs in
Hawaii, Guam and the Commonwealth of the Northern Mariana Islands,
Puerto Rico, United States Virgin Islands, and in the Washington, DC,
Area, March 1994,'' prepared by Runzheimer International under
Government contract OPM-90-0705. This report provides the basis for the
increases in certain cost-of-living allowances (COLA's) being proposed
by OPM in the notice of proposed rulemaking immediately preceding this
notice.

DATES: Consistent with the deadline on comments in response to the
notice of proposed rulemaking immediately preceding this notice, OPM
requests that comments on the report be submitted on or before July 25,
1994.

ADDRESSES: Send or deliver comments to Allan G. Hearne, Methodology
Development Branch, Office of Compensation Policy, Personnel Systems
and Oversight Group, Office of Personnel Management, room 6H31, 1900 E
Street NW., Washington, DC 20415.

FOR FURTHER INFORMATION CONTACT:
Allan G. Hearne, (202) 606-2838.

SUPPLEMENTARY INFORMATION: Under 5 CFR 591.206(c), COLA survey
summaries must be published in the Federal Register. Accordingly, OPM
is publishing the complete ``Report to OPM on Living Costs in Hawaii,
Guam and the Commonwealth of the Northern Mariana Islands, Puerto Rico,
United States Virgin Islands, and the Washington, DC, Area, March
1994,'' produced by Runzheimer International. The Runzheimer report
describes the surveys that were conducted for OPM in the summer of 1993
in Hawaii, Guam, Puerto Rico, the Virgin Islands, and the Washington,
DC, area. It also explains in detail the methodologies, calculations,
and findings.
OPM is publishing the survey results at this time to expedite the
implementation of potential COLA rate adjustments. This notice does not
cover the winter 1994 living-cost surveys conducted in Alaska. The
results of the winter 1994 surveys will be published in a separate
notice later this year after OPM has fully analyzed the results of the
Alaska surveys.
Based on the summer 1993 living-cost surveys, Runzheimer computed
index values of relative living costs in allowance areas using an index
scale where the living costs in the Washington, DC, area are set at
100. (See the Executive summary of the March 1994 Runzheimer report
accompanying this notice.) OPM notes that the summer surveys indicated
that COLA rates in three allowance areas are above levels otherwise
warranted. However, the Treasury, Postal Service, and General
Government Appropriations Act, 1992 (Pub. L. 102-141), bars any
reduction in COLA rates through December 31, 1995. Thus, the only rate
adjustments to be made are rate increases, as described in the notice
of proposed rulemaking immediately preceding this notice.

U.S. Office of Personnel Management.
James B. King,
Director.

Report to OPM on Living Costs In Hawaii, Guam and the Commonwealth of
the Northern Mariana Islands, Puerto Rico, United States Virgin
Islands, and in the Washington, DC Area

March 1994

Table of Contents

Executive Summary
1. Introduction
1.1 Report Objectives
1.2 Changes in This Year's Survey
1.3 Pricing Period
1.4 Living Cost Components
2. Overall Model
2.1 Measurement of Living-Cost Differences
2.1.1 Target Population: Federal Employees
2.1.2 Determination of Expenditure Patterns
2.1.2.1 Source of Expenditure Data
2.1.2.2 Income Level Adjustments
2.1.2.3 Family Size Considerations
2.1.2.4 Analysis of the 1988 Consumer Expenditure Survey
2.2 General Formulae and Applications
2.3 Data Collection Process
2.3.1 In-house Research Staff
2.3.2 Field Researchers--``Research Associates''
2.3.3 On-site Visits by Runzheimer Research Personnel
2.4 Editing and Quality Control Procedures
2.5 Pricing Surveys in Puerto Rico
2.6 Pricing Surveys in Hawaii County
2.7 Surveying the Washington, D.C., Area
3. Consumption Goods & Services
3.1 Component Overview
3.2 Marketbasket Research
3.2.1 Expenditure Research--Category Weightings
3.2.2 Expenditure Research--Subcategory and Item weightings
3.3 Marketbasket Item Specifications
3.3.1 Exchange and Commissary Expenditure Research
3.4 Goods & Services Data Collection Procedures
3.4.1 Data Collection Materials
3.4.2 Outlet Selection
3.4.3 Special Considerations in Guam and Kauai
3.5 Inclusion of Sales and Excise Taxes
3.6 Goods & Services Survey Results
4. Housing
4.1 Component Overview
4.2 Housing Model
4.2.1 Expenditure Research
4.2.2 Development of Housing Profiles
4.2.3 Living Community Selection
4.2.4 Identification and Quantification of Housing-Related
Expenses
4.2.4.1 Utilities
4.2.4.2 Real Estate Taxes
4.2.4.3 Owners/Renters Insurance
4.2.4.4 Maintenance
4.2.4.5 Telephone
4.3 Housing Data Collection Procedures
4.3.1 Homeowner Data Collection
4.3.2 Renter Data Collection
4.4 Housing Analysis
4.4.1 Homeowner Data Analysis
4.4.1.1 Data Trimming
4.4.1.2 Special Considerations
4.4.2 Rental Data Analysis
4.4.2.1 Data Trimming and Special Analyses
4.4.3 Analysis of Housing-Related Expenses
4.5 Housing Survey Results
5. Transportation
5.1 Component Overview
5.2 Private Transportation Methodology
5.2.1 Vehicle Selection and Pricing
5.2.2 Vehicle Trade Cycle
5.2.3 Fuel Performance and Type
5.2.3.1 Impact of Temperature upon Fuel Performance
5.2.3.2 Impact of Road Surface upon Fuel Performance
5.2.3.3 Impact of Gradient upon Fuel Performance
5.2.3.4 Overall Impact upon Fuel Performance
5.2.4 Vehicle Maintenance
5.2.5 Tires
5.2.6 License and Registration Fees, and Miscellaneous Tax
5.2.7 Depreciation
5.2.8 Finance Expense
5.2.9 Vehicle Insurance
5.3 Public Transportation Methodology
5.4 Transportation Survey Results
6. Miscellaneous Expenses
6.1 Component Overview
6.2 Miscellaneous Expense Model
6.2.1 Expenditure Research
6.2.2 Miscellaneous Expense Methodology
6.3 Miscellaneous Expense Data Collection Procedures
6.4 Miscellaneous Expense Survey Results
7. Final Results
7.1 Total Comparative Cost Indexes
7.2 General Comments
7.3 Recommendations

Appendix

Appendix 1 Consumer Expenditure Survey (CES)
Appendix 2 Marketbasket Descriptions
Appendix 3 Pricing Changes: Good & Services/Miscellaneous Expense/
Transportation/Housing Related
Appendix 4 Consumption Goods & Services Analysis
Appendix 5 Nonforeign Area Cost-of-Living Allowances Price Survey
Data and Background Survey Data Collection Procedures
Appendix 6 1992/1993 OPM Living Community Selection
Appendix 7 Housing Cost Analysis
Appendix 8 Housing Analysis
Appendix 9A Analysis of Home Sales Data
Appendix 9B Analysis of Rental Data
Appendix 10A Private Transportation Cost Analysis
Appendix 10B Auto Insurance Calculation Worksheet
Appendix 11 Transportation Analysis
Appendix 12 Miscellaneous Expense Analysis
Appendix 12 Multiple Survey Areas
Appendix 14 Component Expenditure Amounts & Total Comparative Cost
Indexes

Executive Summary

This report culminates the fourth living-cost comparison study
undertaken by Runzheimer International for the Office of Personnel
Management (OPM) under contract OPM-90-0705. The contract requires
Runzheimer to:
(1) Survey living costs in 7 cost-of-living allowance (COLA) areas
and the Washington, DC, area, and
(2) Compare living costs between the areas and the DC area.
To determine living costs in the identified areas and build this
report, Runzheimer researched over 3,000 outlets and gathered more than
12,000 price quotes.
This report presents the results of the living-cost surveys
conducted during the summer of 1993. The results of the winter 1994
living-cost surveys will be presented in a subsequent report to be
provided to OPM in mid-1994.
To ease interpretation of the research results, we display the
outcome of the comparisons as living-cost indexes in the table below.
In addition, the table shows living-cost indexes for federal employees
who have unlimited access to commissary and exchange facilities because
of their employment. The index for the Washington, DC, area (not shown)
is 100.00 because it is, by definition, the reference area.

Final Cost Comparison Indexes
------------------------------------------------------------------------
Local Commissary
Allowance area pricing & exchange
------------------------------------------------------------------------
City & Cnty of Honolulu, Hawaii............... 122.90 120.26
Hawaii Cnty, Hawaii........................... 109.63 NA
Kauai Cnty, Hawaii............................ 119.27 NA
Maui Cnty, Hawaii............................. 119.32 NA
Guam, CNMI*................................... 122.25 120.81
Puerto Rico................................... 103.00 102.17
U.S. Virgin Islands........................... 117.81 NA
------------------------------------------------------------------------
*Commonwealth of the Northern Mariana Islands.
NA=Not Applicable.

One of the changes OPM requested this year was to separate the
reporting of the summer and winter living-cost surveys. As noted
earlier, this has been done. The purpose of this change was to allow
OPM to adjust COLA where warranted in a more timely manner.
Another change was to combine the St. Croix and St. Thomas, Virgin
Island, data to form a single index for the U.S. Virgin Islands. This
change was made to address concerns about the quantity of data,
obtained in the Virgin Islands. Combining the two areas significantly
increases the data base and improves price comparisons.
For the 1993 summer surveys, OPM asked Runzheimer to change of its
data collection procedures and data analyses relative to last year's
surveys. Runzheimer also initiated other changes with OPM's approval.
Some of these changes included:
Using a moving average approach to introduce new Federal
employment weights;
Increasing the quantity of housing data obtained in
certain areas--most notably parts of the Virgin Islands;
Making minor changes in pricing sources for certain items
to refine the comparisons of D.C. and allowance area prices;
Taking into account the effects of the earthquake in Guam
by ensuring that goods & services pricing--most notably fresh produce--
was not abnormally skewed;
Taking into account the effects of Hurricane Iniki in
Kauai by ensuring that prices--particularly housing costs--were not
abnormally skewed;
Employing a new methodology for collecting and analyzing
Virgin Island, Guam, and Puerto Rico automobile insurance rates; and
Employing a new methodology for determining automobile
maintenance costs.
In addition, to monitor, fine-tune, and maintain effective control
of the data-gathering efforts in both the Pacific and Caribbean
regions, two of Runzheimer's senior research staff traveled to these
regions to visit retail outlets, Runzheimer research associates,
housing data sources, and living communities.
We discuss these and other adjustments in appropriate sections
throughout this report.

Report to OPM on Living Costs in Hawaii, Guam and the Commonwealth of
the Northern Mariana Islands, Puerto Rico, U.S. Virgin Islands, and in
the Washington, DC Area

1. Introduction

1.1 Report Objectives

This comprehensive report culminates data-gathering and research
work undertaken in 1993 as required by Task 2 of contract OPM-90-0705
between the Office of Personnel Management (OPM) and Runzheimer
International. The report details the results of Runzheimer
International's surveys of over 3,000 outlets to obtain more than
12,000 price quotes and the analyses of the data.
This is the fourth report Runzheimer has produced for OPM under
this contract. In 1990, in fulfillment of Task 1 of the contract,
Runzheimer worked with OPM to design a model for estimating comparative
living costs between the allowance areas and the Washington, D.C.,
area. Task 2 of the contract required that Runzheimer apply the model
by conducting living-cost surveys, analyzing the results, and
developing living-cost comparative indexes. On February 26, 1991, OPM
published that model and the results of the first surveys conducted
under the model in the Federal Register. On December 10, 1992, OPM
published in the Federal Register the second report, which covered the
summer 1991 and winter 1992 surveys. On August 30, 1993, OPM published
in the Federal Register the next report that covered the summer 1992
and winter 1993 surveys.
Unlike previous reports, this report provides only the results of
the summer 1993 surveys. This change was made to allow OPM the
opportunity to adjust COLA rates where warranted in a more timely
manner. Results of the 1994 winter surveys (i.e., the Alaska surveys)
will be presented in a separate report to be provided to OPM in mid-
1994.
The analyses in this report establish the comparative cost
differences between the allowance areas listed below and the
Washington, D.C., area. By law, Washington, D.C., is the base of
``reference'' area for the nonforeign-area COLA program.

1. City and County of Honolulu, Hawaii
2. Hawaii County, Hawaii
3. Kauai County, Hawaii
4. Maui County, Hawaii
5. Guam and the Commonwealth of the Northern Mariana Islands (CNMI)
6. Puerto Rico
7. U.S. Virgin Islands

Under OPM regulations, federal civilian employees who have
unlimited access to commissaries and post exchanges due to their
employment by the government may receive a different allowance rate
than other federal employees. This regulation does not apply to federal
employees who have limited access or unlimited access for other
reasons--e.g., being married to active or retired military personnel.
Task 2 of the OPM contract also required Runzheimer International
to calculate comparative living costs in the areas listed below and the
Washington, D.C., area for federal civilian employees who have access
to military commissaries and post exchanges.

1. City and County of Honolulu, HI
2. Guam/CNMI
3. Puerto Rico

1.2 Changes in This Year's Survey

Runzheimer and OPM made several changes to the surveys and
analyses, including--
reporting the results of the summer and winter surveys
separately;
using a moving average approach to introduce new Federal
employment weights;
combining St. Croix and St. Thomas data to produce indexes
for the Virgin Islands as a whole;
increasing the sample size of housing information in
several areas, including Hawaii, San Juan, St. Croix, and the Maryland
suburbs of the Washington, D.C., area;
developing a percent-to-market value formula to calculate
real estate taxes for San Juan because tax assessment data were no
longer available as a result of a decentralization of property tax
function in Puerto Rico;
increasing the comparability of automobile insurance in
the Virgin Islands, Guam, and Puerto Rico, compared with the
Washington, D.C., area;
surveying only the cash price at branded gas stations
unless only non-branded stations are available; and
more accurately defining the distinguishing differences
between family dining and fine dining.
Runzheimer has continued to include catalog sales in its survey.
Since the Sears catalogs have been discontinued. Runzheimer has
researched hundreds of catalogs to determine which are most
appropriate. Runzheimer researchers found that most catalogs have
uniform shipping charges. Only catalogs that sell merchandise in the
allowance areas and the Washington, D.C., area were used.
Appendix 3 identifies Goods & Services, Miscellaneous Expense, and
Housing Related pricing changes. Current housing data can be found in
appendix 9A and Appendix 9B. Other changes are discussed where
applicable in the report.

1.3 Pricing Period

Consistent with last year's tropical-area surveys. Runzheimer
collected data for the Hawaii, Guam/CNMI, Puerto Rico, and Virgin
Islands allowance areas (and the Washington, D.C., area) in August,
September, and October of 1993, pricing most items during August. Also
during August, our research associates priced durable goods, such as
cars, and, in October, items such as homeowner insurance, which depend
on the pricing of other items (i.e., housing).
To ensure consistent seasonable catalog pricing, Runzehimer used
spring/summer catalogs for the catalog items covered in these surveys.

1.4 Living Cost Components

In accordance with federal regulations, expense components
Runzheimer costed to develop analyses, comparisons, and the report
were:

1. Housing and Housing Related Expenses
2. Transportation
3. Consumption Goods & Services
4. Miscellaneous Expenses

Runzehimer factored sales, excise and property taxes into the
analysis where applicable. However, in keeping with previous reports,
we did not factor federal, state and local income taxes into the
analysis. Because income taxes significantly affect living-cost
analyses, Runzheimer and OPM are researching the issue of including
income taxes in future surveys.
Educational opportunities vary significantly among locations in
terms of availability, quality, and other factors. Runzheimer analyst
and OPM officials agree that, without additional information, attempts
to measure cost differences in education in the selected areas would be
highly subjective and would not add to the integrity of the model.
Therefore, education expense is not included in the model or surveys.

2. Overall Model

2.1 Measurement of Living-Cost Differences

The most common and most widely accepted way to measure living-cost
differences between and among locations is to select representative
items that people purchase in these locations and to calculate the
respective cost differences, combining them according to their
importance to one another (as measured by relative percentage of
expenditures). Runzheimer applied this methodology to compare the
living costs in each of the allowance areas with the living costs in
the Washington, D.C. area.
To move from this basic concept to computing comparative living
costs between each allowance area and the Washington, D.C., area,
Runzheimer followed five main processes or steps:
Step 1 Identify the segment of the population for which this
analysis is being targeted (i.e., the target population).
Step 2 Determine how these people spend their money.
Step 3 Select items to represent the expense categories for which
these people spend their money.
Step 4 Conduct pricing surveys of the selected items in each area.
Step 5 Analyze cost ratios for the selected items and aggregate
them according to the relative importance of each item.
2.1.1 Target Population: Federal Employees
Runzheimer's living-cost model measures living-cost differences for
non-military Federal employees having annual base salaries between
$10,000 and $80,000, the salary range of the 1990 General Schedule (GS)
of the Federal Government. Because living-cost differences may vary
depending on an employee's income level, Runzheimer designed its
analytical model to identify living costs at three income levels.
In its first report to OPM, Runzheimer used the salary distribution
of all General Schedule employees as of March 31, 1990, which OPM
supplied, to determine the income levels that most accurately represent
the Federal employee population. After analyzing the array of salary
data, Runzheimer picked the midpoints of the lower, middle and upper
thirds of the distribution as its three income levels ($18,000, $28,400
and $45,200 respectively).
Runzheimer applied the same income levels for this report as it did
for the first. In previous reports, Runzheimer recommended that OPM
consider introducing changes in income levels and weights on a gradual
basis. OPM agreed, and this year OPM introduced new Federal employment
weights that are based on a moving average. OPM has informed Runzheimer
that OPM plans to introduce other changes, such as in the
representative income levels and Consumer Expenditure Survey (CES)
weights, in future surveys.
Runzheimer uses Federal employment weights in the model in two
ways: (1) to combine survey data from multiple survey areas within a
single allowance area and (2) to combine relative living costs by
income level within each allowance area into a single index for the
area (as required by section 591.205(c) of title 5, Code of Federal
regulations).
OPM's moving average allows the gradual introduction of new
employment distribution data over time. The weights are based on a
three-year average of GS employment. Each year, the latest GS
employment data will be added to the three-year average, the oldest
data will be deleted, and a new three-year average will be computed.
This will keep the weights current while mitigating any fluctuations
due to short-term changes in Federal employment.
In this first application of the moving average, OPM is departing
slightly from the process described above in that the three employment
distributions used are for the periods of 1990, 1992, and 1993. The
1990 rather than 1991 employment distribution is used because the model
previously used 1990 data only. (See Appendix 13.)
2.1.2 Determination of Expenditure Patterns
2.1.2.1 Source of Expenditure Data
Conforming with last year's process, Runzheimer used the ``prepub''
statistical reports from the 1988 CES dated February 13, 1990 (see
Appendix 1) as the basis for weighting expenditure patterns.
2.1.2.2 Income Level Adjustments
Because the CES reflected 1988 expenditure levels, Runzheimer
reduced the three 1990 incomes back to 1988 levels before beginning the
expenditure analysis. To calculate estimated 1988 income levels,
Runzheimer used the average percentage salary increases of Federal
employees for the two-year period in question as supplied by OPM
officials (4.1% increase 1988-89 and 3.6% increase 1989-90, resulting
in a 7.85% two-year increase). This adjustment reduced the 1990 income
levels to estimated 1988 levels of $16,700, $26,300, and $41,900.
2.1.2.3 Family Size Considerations
A family size of 2.6 was inherent in the weighting scheme
Runzheimer employed to price all allowance areas. Derived from CES
research, the number represented an average for the nation.
2.1.2.4 Analysis of the 1988 Consumer Expenditure Survey
From the 1988 CES, Runzheimer used the statistical report entitled
``Table 2. Income before Taxes,'' which listed average expenditures for
families earning similar incomes, organized into eight income ranges.
Runzheimer analyzed these data to develop typical spending patterns for
the three income profiles identified in 2.1.2.2. (The table below
displays the results of the analysis.)
Seven income categories encompassed these three income profiles:

$10,000 to $14,999
$15,000 to $19,999
$20,000 to $29,999
$30,000 to $39,999
$40,000 to $49,999
$50,000 and over
All respondents combined

The 1988 CES grouped expenses into small, logical families of
items. For example, the report divided money spent by families on beef
into four groups: ground beef, roast, steak and other beef. The steak
and roast groupings were further separated into smaller clusters of
items (e.g., sirloin and round steak, chuck and round roast).
Drawing on this survey of expenditure data, Runzheimer sorted the
item groupings into the four main cost components specified in the OPM
regulations: Consumption Goods & Services, Transportation, Housing, and
Miscellaneous Expenses. Runzheimer observed that families in the lower
income ranges spent more of their money, as a percentage of total
expenses, on goods and services and housing than families in higher
income ranges. Also, families spent approximately the same percentage
of their total expenses on transportation, regardless of income.
Consequently, the Miscellaneous Expense component--which includes such
things as medical-care expenses, contributions, gifts to non-family
members, pension funds, long-term savings and investments, and life
insurance premiums--increased as a percentage of total expenses as
income increased.
To develop accurate and defensible weighting patterns for the three
income levels, Runzheimer performed linear regression analysis on the
selected 1988 CES data. Listed below are the results of Runzheimer's
analysis for the income ranges listed on the preceding page:

Component Expenses Expressed as a Percentage of Total Expenses
----------------------------------------------------------------------------------------------------------------
Income Goods &
Income Level 1991 Level 1988 Services Housing Transportation Misc. Total
(Est.) (percent) (percent) (percent) (percent) (percent)
----------------------------------------------------------------------------------------------------------------
$18,000........................ $16,700 39.59 24.35 20.76 15.30 100.00
28,400......................... 26,300 39.15 23.48 20.33 17.04 100.00
45,200......................... 41,900 38.74 22.66 19.94 18.66 100.00
----------------------------------------------------------------------------------------------------------------

Runzheimer further sorted Goods & Services into ten categories and
used linear regression techniques to provide ratios of renters to
homeowners at each income level. Statistics on these component
groupings appear later in this report.

2.2 General Formulae and Applications

An ```index'' is a mathematical way to compare one price (or set of
data) with another. For example, if a price index for a can of green
beans is 110, that means that a can of green beans costs 10% more in
the pricing area (i.e., allowance area) than in the reference area
(i.e., the Washington, D.C., area).
Runzheimer computed indexes for hundreds of items. To combine these
indexes, Runzheimer applied weights from the CES that reflected the
relative amount consumers normally spend on the items. For example, the
price of a can of green beans has a lower weight than the price of a
pound of apples because, according to the CES, people generally spend
less on canned green beans than on apples.
Runzheimer employed an indexing methodology known as Laspeyres to
derive total cost indexes for each of three income levels and for each
costed location. As applied to living-cost research, the Laspeyres
index reflects the expenditure patterns of the people in the reference
area (i.e., the Washington, D.C., area) to weight the prices. Because
detailed CES data by income level are not published for the Washington,
D.C., area, Runzheimer used nationwide CES data to compute weights.
Consequently, Runzheimer did not technically apply the Laspeyres index
methodology in its pure form. Nevertheless, Runzheimer firmly believes
that this nuance does not invalidate the price comparisons presented in
this report.
As described in the example above and in greater detail in sections
3.2 and 6.2, Runzheimer applied the Laspeyres methodology to compute
price indexes for the Goods & Services and the Miscellaneous Expenses
components, respectively. For the Transportation and the Housing
components, Runzheimer used a combination of a cost-build-up approach
and the Laspeyres methodology to compute component indexes.
In conformance with section 591.205(c), title 5, Code of Federal
Regulations, Runzheimer followed a five-step process to derive the
overall total indexes for each allowance area. First, Runzheimer used
the CES data and the income ranges described in section 2.1.2.4 above
to derive the amount of money consumers typically spend on each
component at each income level. These amounts appear in the table below
and in Appendix 14.

Typical Consumer Expenditures by Income Level and Component
--------------------------------------------------------------------------------------------------------------------------------------------------------
Income level Goods & services Own/rent Transportation Misc. Total
--------------------------------------------------------------------------------------------------------------------------------------------------------
Lower.................................................... $7,126 $4,383 $3,737 $2,754 $18,000
Middle................................................... 11,119 6,668 5,774 4,839 28,400
Upper.................................................... 17,510 10,242 9,013 8,434 45,200
--------------------------------------------------------------------------------------------------------------------------------------------------------
Note: Values may not total because of rounding.

Second, for each allowance area, Runzheimer multiplied the values
above by the component indexes for the allowance area. Because the
Housing component consisted of two indexes (one for owners and another
for renters), we produced two sets of total relative costs--one for
owners and another for renters.
Third, for each allowance area and income level, Runzheimer
combined the total relative costs for owners and renters using the
proportion of owners and renters as identified in the CES to weight the
costs. (See section 4.2.1) This produced an overall, average, relative
living cost at each income level in each allowance area.
Fourth, for each allowance area, the overall, average relative
costs by income level were combined using federal employment weights
based on the employment at each income level in the allowance area.
Applying the same allowance-area employment weights, Runzheimer
computed an overall, average cost for the reference area.
The last step was to divide the overall, average relative cost for
the allowance area by the overall, average cost for the reference area
to produce the final index. (See Appendix 14 for the calculations for
each index.)

2.3 Data Collection Process

As noted earlier, Runzheimer obtained price information on over
12,000 items from over 3,000 outlets. To accomplish this important
research effort, we selected the most efficient and effective
information-gathering approaches possible. This section describes the
various approaches.
2.3.1 In-house Research Staff
Runzheimer research personnel at its corporate headquarters in
Rochester, Wisconsin, played a major role in all data-collection
activities. These professionals:
Contacted manufacturers, trade associations, governmental
agencies, and retail establishments to ensure that suitable items were
selected and priced;
Contacted professionals in the real estate business in
each of the costed locations to obtain general information as well as
specific rental rates and home market values;
Conducted pricing surveys on site and by telephone for
many items;
Served as a liaison for field researchers;
Performed hundreds of quality control checks in
conformance with editing rules communicated by Runzheimer to OPM once
the data had been collected (these checks often involved verification
of the survey data through telephone calls as well as comparing current
data-gathering results with those from the 1992-1993 survey); and
Analyzed and computed the category, component and total
comparative cost indexes.
2.3.2 Field Researchers--``Research Associates''
Collection of most price data was best accomplished through
personal visits to retail outlets (e.g., grocery, clothing,
automobiles). For these activities, Runzheimer hired residents of each
allowance area as independent contractors (``research associates'').
For years, when measuring living costs for its clients, Runzheimer has
applied this approach to data collection in over 80 countries
worldwide.
To avoid any real or perceived conflicts of interest, Runzheimer
did not hire persons as research associates who were either employees
of the federal government, or who had immediate family who were
employees of the federal government.
2.3.3 On-site Visits by Runzheimer Research Personnel
Full-time Runzheimer research professionals travelled to selected
allowance areas to supervise data-collection activities and perform
various quality-control checks on the data as necessary. Each such
visit occurred during the pricing period.
The researchers visited living communities within the allowance
areas to look at housing accommodations personally and to talk with
local real estate professionals. They also visited numerous retail
outlets to verify item quality, selection and price levels in general.
In addition, these researches met with Runzheimer's research
associate(s) to answer any data-collection questions and to provide any
additional training and instruction as necessary.

2.4 Editing and Quality Control Procedures

Runzheimer's experience in measuring living-cost differences
enabled us to establish editing and quality-control procedures at all
stages of collecting and analyzing data. All data provided by research
associates were manually reviewed by analysts prior to being entered
into Runzheimer's computer system. Data elements were subsequently
checked through software programs.
Federal regulations in section 591.205(b)(1)(i), title 5, Code of
Federal Regulations, state that, ``Whenever possible, exact brands and
models are priced in each location.'' Every effort was made to satisfy
this objective. (See section 3.3 for a discussion of brand and model
selection.) Nevertheless, in a number of the allowance areas, the exact
brands and models were either not readily available or not available at
all. In these instances, editing decisions and substitutions were
needed.
Runzheimer defines ``editing'' as the removal and/or replacement of
a price quote based on consistent and logical criteria. In all areas,
Runzheimer was concerned that items of lesser or greater quality than
the item specified might inadvertently be included in the analysis and
bias the results. Therefore, any price quote that varied significantly
from other price quotes for the item was flagged, verified, and if
necessary, eliminated from the analysis.
Removing an item from a location analysis causes redistribution of
its weight to other items in its subcategory (or category when no
subcategory exists). Consequently, whenever possible, Runzheimer
avoided removing an unpriced item from a location analysis. When the
review process revealed a missing price for an item, Runzheimer
resurveyed to obtain a price wherever possible.

2.5 Pricing Surveys in Hawaii County, Puerto Rico, and the Virgin
Islands

Three allowances areas have multiple survey areas within each
allowance areas. The three allowance areas are Hawaii County, Hawaii;
Puerto Rico; and the U.S. Virgin Islands. For each of these areas, OPM
provided Runzheimer GS employment distribution data based on the
moving-average approach described in section 2.1.1.
In the Hawaii County allowance area, two areas are surveyed: Hilo
and Kailua Kona. The OPM data indicated that approximately 82% of the
GS employees worked in the Hilo area, and the remaining 18% worked in
the Kailua Kona area. To combine prices from both cities, Runzheimer
used approximately an 82% Hilo and 18% Kailua Kona weighting. (See
Appendix 13.)
Similarly, in the Puerto Rico allowance area, two areas are
surveyed: Mayaguez and San Juan. The OPM data indicated that
approximately 84% of the GS workforce in Puerto Rico work at facilities
within or near the San Juan-Caguas-Areciro Consolidated Metropolitan
Statistical Area. The remaining 16% is distributed throughout Puerto
Rico without any particular concentration. However, more of these
remaining employees generally appeared to live closer to Mayaguez than
any other large Puerto Rican city. Therefore, with OPM's approval,
Mayaguez was chosen for the survey. To combine the prices from both
cities, Runzheimer used approximately an 84% San Juan and 16% Mayaguez
weighting.
As noted earlier, this year OPM asked Runzheimer to combine survey
data from St. Croix and St. Thomas, Virgin Islands. The same approach
used in Hawaii County and Puerto Rico was applied to the Virgin Islands
data. OPM's data indicated that approximately 46% of the GS employees
worked on St. Croix, and that the remaining 54% worked on St. Thomas
and St. John. Therefore, to combine the prices from these areas,
Runzheimer used approximately a 46% St. Croix and 54% St. Thomas/St.
John weighting.

2.6 Surveying the Washington, DC, Area

OPM defined the Washington, DC., area in the federal regulations as
the Washington DC-MD-VA Metropolitan Statistical Area. Because federal
employees who work in this area reside in Virginia, in Maryland, and in
the District of Columbia, Runzheimer selected retail outlets and living
communities from all three. Runzheimer's model gave equal weight to the
average prices in each geographic area.
Because of the size and diversity of the Washington, D.C., area,
Runzheimer conducted substantially more pricing surveys there than in
other areas. For the Goods & Services component, Runzheimer generally
surveyed to obtain six times as many price quotes in the Washington,
D.C., area as in the typical allowance area. For the Housing,
Transportation and Miscellaneous Expense components, data collection
was generally triple that of the typical allowance area.

3. Consumption Goods & Services

3.1 Component Overview

Based on the CES data, the Goods & Services component consisted of
ten categories of family expense:

Food at Home
Food Away from Home
Tobacco
Alcohol
Furnishings & Household Operations
Clothing
Domestic Services
Professional Services
Personal Care
Recreation

To aid in quality control and analysis of future pricings,
Runzheimer further subdivided four of the largest categories--food at
home, furnishings and household operations, clothing, and recreation--
into subcategories. Specific examples of products and services from
these four subdivided categories can be found below. These examples
only represent a minor portion of the total number of items (products
and services) Runzheimer priced within each subdivided category. (See
Appendix 2 for a complete description of all marketbasket items.)

Examples of Subcategories and Items Surveyed in the Four Major Goods &
Services Categories
Goods and Subcategories and items surveyed (examples)
services
category

------------------------------------------------------------------------
Food at Home.... Meats Cereals & Breads Groceries.
Pork Chops Cookies Coffee.
Whole Chicken Spaghetti Ketchup.
Ground Beef Cake Margarine.
Dairy Fruits &
Vegetables
Milk Apples
Cheddar Cheese Frozen Peas
Fresh Oranges
Furnishing and Services Furniture Misc. Household
Household Eqpt.
Operations.
Appliance Repair Living Room Chair
Supplies Major Appliances Hammer.
Toilet Tissue Kitchen Range Electric Drill.
Laundry Soap Refrigerator Lawn Trimmer.
Household Housewares &
Textiles
Bath Towel Small Appliance
Two-slice Toaster
Clothing........ Men's and Boy's Infant's
Boy's Jeans Disposable
Diapers
Man's Jeans Footwear
Women's and Man's Shoes
Girl's
Woman's Slacks Apparel Products
and Services
Girl's Blouse Coin Laundry
Girl's Jeans
Recreation...... Fees and TV, Radio and Entertainment.
Admissions Eqpt
Bowling Video Rental Board game.
Golf Pets Reading.
Pet Food Magazine.

From its ten categories of expense (which include the four
subdivided categories above), Runzheimer selected a marketbasket of
items on which to base its goods and services analysis. A
``marketbasket'' is a selected group of products and services that
represent hundred or even thousands of other items. Pricing every item
available to consumers in a given locale would be unnecessary and
inefficient.
Runzheimer selected typically purchased items and weighted these
according to their relative importance in terms of consumer expenditure
patterns. Each marketbasket item represented a specific group of
related expense items. Using CES data, we determined the relative
importance (weight) of each item. We compared the average price of each
marketbasket item in each allowance area with the average price in the
Washington, D.C., area. The price differences (expressed as indexes)
were aggregated based on the item, subcategory and category weighting,
resulting in a total Goods & Services component index at each income
level.
In 1991, OPM directed Runzheimer to include catalog pricing
(including applicable shipping costs) to reflect this common purchasing
option in allowance areas. OPM identified items to survey based on
comments received on the 1990 survey. Runzheimer identified items to
survey that were either unavailable or difficult to find in each
allowance area. Together, OPM and Runzheimer agreed to survey five
marketbasket items by catalog: For the 1992-1993 survey, OPM asked
Runzheimer to survey additional items by catalog; for example,
furniture. For this year's survey, Runzheimer has added additional
catalog items. For the 1992-1993 survey, Runzheimer found that the cost
to ship certain catalog items to Guam varied between catalog outlets.
With OPM's approval Runzheimer continued to price a private freight
company that specializes in shipments to Guam for shipping costs.
In each survey area, Runzheimer generally requested three price
quotes for each item (and sometimes more than three) from the local
economy--one from each of three different outlets. It should be
remembered that Hawaii County, Hawaii; Puerto Rico; and the Virgin
Islands each have two separate survey areas. Therefore, Runzheimer
generally doubled the number of price quotes obtained in these
allowance areas.

3.2 Marketbasket Research

3.2.1 Expenditure Research--Category Weightings
Ruzheimer tabulated the expense data from the 1988 Consumer
Expenditure Survey according to the ten categories of goods and
services. As in the component analysis, Runzheimer used the expense
data from the seven most appropriate income ranges as input into a
linear regression analysis. From that analysis, Runzheimer calculated
the category weightings for each income level as listed belows:

Category Weightings Expressed as a Percentage
------------------------------------------------------------------------
Category Lower Middle Upper
------------------------------------------------------------------------
Food at Home.............................. 25.52 22.38 19.35
Food Away from Home....................... 15.95 16.09 16.23
Tobacco................................... 3.13 2.54 1.96
Alcohol................................... 2.92 2.79 2.67
Furnishings & Hsld. Op.................... 14.35 15.95 17.49
Clothing.................................. 14.24 14.93 15.59
Domestic Service.......................... 1.78 1.79 1.81
Professional Services..................... 5.77 5.84 5.91
Personal Care............................. 3.57 3.47 3.38
Recreation................................ 12.77 14.22 15.61
-----------------------------
Totals.................................... 100.00 100.00 100.00
------------------------------------------------------------------------

3.2.2 Expenditure Research--Subcategory and Item Weightings
Runzheimer also drew upon the expense data from the 1988 CES to
determine proper subcategory and item weightings and to identify
marketbasket items. Logical groupings of family expenditures provided
the basis for subcategory and item weights. Unlike category weightings,
which vary by income level, subcategory and item weightings are
computed from national aggregate expenditures only (i.e., all three
income levels used the same set of subcategory and item weightings) as
this approach is most common in similar public and private sector cost-
of-living analyses.
Runzheimer's expenditure research process included procedures to
ensure that no marketbasket item had an overwhelmingly large or
insignificantly small item weighting.

3.3 Marketbasket Item Specifications

From each logical expense grouping, Runzheimer selected one or more
marketbasket items to represent all items in the grouping. When
selecting specific items for the marketbasket, Runzheimer worked to
satisfy these three criteria:
Items should be readily available in all locations if
possible or should be items of local significance.
Item price levels should logically represent the price
levels of unselected items in the ``logical grouping.''
Items should have the same or nearly the same application
in all locations.
Appendix 2 lists Runzheimer's marketbasket items. Once an item was
selected, Runzheimer's research analysts identified the specific brand
and/or model/size of each item available in all (or most) locations.
For some items, this involved contacting manufacturers, trade
associations, retail establishments, etc. For other items, isolating
specifications was quite straight forward because of their nature
(e.g., bread, nonprescription pain reliever). Appendix 3 identifies
changes in the current items selected for pricing in the Goods &
Services, Miscellaneous Expense, and Housing Related categories, along
with explanation for the changes.
3.3.1 Exchange and Commissary Expenditure Research
Runzheimer used the same marketbasket items to price commissaries
and exchanges as were used for the local pricings. We obtained one
price quote for each marketbasket item surveyed in these facilities.
Runzheimer did not assume that people with access to military
facilities made all purchases in these facilities. Instead, we used
OPM's 1980 Living Pattern Survey of federal employees to determine the
percentage of purchases that families typically make in military
facilities versus local outlets. For example, as the following table
shows, in Honolulu, 52.7% of Food at Home is purchased at a PX or
commissary. These percentages were used to aggregate the local and
commissary/exchange prices into one set of appropriate blended prices.
(The blended prices were compared to the local prices in the
Washington, DC, areas just as each allowance area's local prices were.)

Percentages of Purchases Made at PX/Commissaries
------------------------------------------------------------------------
Allowance Areas
Category --------------------------------
Guam Honolulu San Juan
------------------------------------------------------------------------
Food at Home........................... 70.0 52.7 29.1
Food Away.............................. 0.0 0.0 0.0
Tobacco................................ 64.0 84.0 65.0
Alcohol................................ 76.0 73.0 64.0
Furn. & Hsld. Op....................... 64.5 44.2 32.7
Clothing............................... 43.7 34.0 10.7
Domestic Service....................... 0.0 0.0 0.0
Professional Services.................. 0.0 0.0 0.0
Personal Care.......................... 49.3 33.3 7.3
Recreation............................. 49.7 32.7 15.7
------------------------------------------------------------------------

3.4 Goods & Services Data Collection Procedures

3.4.1 Data Collection Materials
The living-cost surveys conform with the provisions of the
Paperwork Reduction Act and are approved by the Office of Management
and Budget (OMB).
Runzheimer collected data with OMB-approved data collection
materials (see Appendix 5). All Runzheimer-developed worksheets
conformed to the OMB-approved materials.
3.4.2 Outlet Selection
Proper outlet selection is crucial to measuring living-cost
differences accurately because misjudgment can seriously affect survey
results. Runzheimer paid particular attention to choosing appropriate
outlets, focusing on three key guidelines to ensure proper outlet
selection.
First, for areas that had numerous outlets from which to choose,
Runzheimer identified targets in several different geographic areas.
For example, in the Washington, D.C., area, Runzheimer selected outlets
in and around six different geographic areas: that is, two areas in
Virginia, two in Maryland and two in the District of Columbia.
Runzheimer's second guideline was that for any one marketbasket
item, all outlets be similar in type. For example, wherever possible,
Runzheimer surveyed the prices of blue jeans at department stores,
hammers at hardware stores, and refrigerators at appliance stores.
Gathering prices for the same item from hardware stores in one area and
discount stores in another could have distorted price comparisons.
The last guideline involved the diversity of outlets in
Runzheimer's sample. We believe that pricing different items in
different types of outlets more accurately portrays living-cost
differences. For example, for efficiency, Runzheimer could have priced
all clothing items in department stores. However, to incorporate price
levels at other types of outlets that sell clothing items, whenever
possible, Runzheimer surveyed some items in men's and women's clothing
stores, some items in department stores, other items in shoe stores and
still other items in discount department stores.
Runzheimer's research analysts selected outlets on the basis of:
Personal experience of Runzheimer on-site research
associates and travelling researchers;
Informal telephone interviews with knowledgeable residents
in each area;
Yellow pages sections of area telephone books;
Area chambers of commerce and information bureaus; and
Experience gained from other surveys conducted by
Runzheimer.
Runzheimer obtains outlets from various sources because no one
source lists all outlets available. For example, in the allowance areas
especially, many outlets choose not to advertise in the yellow pages.
With new businesses constantly appearing (and old ones
disappearing), outlet selection will be an ongoing process. Also, one
can expect a portion of outlets to refuse to participate every year.
Therefore, updating Runzheimer's outlet sample is a necessary and
important part of each pricing survey.
An example of refining outlet selection for this year's survey is
the attention paid to defining the distinguishing differences between
restaurants categorized as appropriate for family dining and those
appropriate for fine dining. OPM developed a matrix which defined such
characteristics as menu selections, atmosphere, table setting, seating,
reservations, and AAA ratings. This allowed Runzheimer researchers to
compare apples to apples between the allowance areas and the
Washington, D.C., area. Also, OPM established guidelines directing
Runzheimer to survey 100% family restaurants for breakfast, roughly 75%
for lunch, and roughly 66% for dinner.
3.4.3 Special Considerations in Guam and Kauai
The effects of Typhoon Omar and Hurricane Iniki in 1992 on the
economies of Guam and Kauai, respectively, has been minimal over the
long term. At the time, Runzheimer officials worked with local research
associates to determine the impact of the storm on local prices.
This is not to say that the storms had no effect on the local
economies. Indeed, prices in these two areas for the 1993-1994 survey
have been re-examined carefully.
Of equal concern this year was the earthquake that rocked Guam
during August, causing considerable damage and seriously disrupting the
flow of fresh produce into the island. For a period of about two weeks,
all such produce had to be flown in at considerably higher cost.
Fortunately, when normal shipping channels reopened, ``normal'' pricing
was quickly restored, causing negligible affect on the Goods & Services
pricing. Also, a senior Runzheimer researcher from the headquarters
office traveled to Guam immediately following the earthquake to
determine the full extent of the impact on goods and services pricing.

3.5 Inclusion of Sales and Excise Taxes

For all items subject to sales tax, the appropriate amount of tax
was added prior to analysis. Runzheimer also included all applicable
sales, property and excise taxes to items that were part of other
living-cost components. For example, automobile purchase costs in
Puerto Rico include an excise tax based upon vehicle type and dealer's
acquisition cost. Runzheimer included this tax in computing the total
purchase price of a vehicle in Puerto Rico. (See sections 5.2.1 and
5.2.7.)
Runzheimer gathered applicable information on taxes by contacting
appropriate sources of information in the allowance areas, such as the
Excise Tax Officer of the St. Thomas Bureau of Internal Revenue.
Runzheimer also drew upon appropriate tax publications, such as the
State of Maryland's Sales and Use Tax Laws and Regulations and the
``General Excise Tax Law'' (Chapter 237) of the Hawaii Tax Reports.

3.6 Goods & Services Survey Results

In section 2.2 of this report, Runzheimer presented a detailed
explanation of the economic model used to analyze the price data. As it
applies to Goods & Services, the approach involved comparing the
average prices of marketbasket items in each allowance area with those
in the Washington, D.C., area. The resulting price ratios were
aggregated into subcategory and then category indexes using the
expenditure weightings derived from the 1988 CES.
In the area of professional services (accounting and legal fees),
Runzheimer and OPM noted the existence of a relatively few extreme
values that were either exceptionally low or high cost. In situations
such as this, statisticians frequently use the median or trim the data
in some manner to enhance reliability. Consequently, Runzheimer
recommended that the observations be ranked from low to high and that
the top and bottom 20% of the observations be ``trimmed'' (i.e.,
eliminated) from the data before averages or trends were calculated.
(Data were not trimmed if there were four or fewer observations.) OPM
agreed. These procedures reduce the influence of anomalies and make
survey results more stable from one year to the next.
Appendix 4 contains tables showing the ten category indexes, the
three weighting patterns, and the three total consumption Goods &
Services indexes for each allowance area. The Washington, D.C., area
does not require a table because it is, by definition, ``the reference
location'' where all category and component indexes equal 100.

4. Housing

4.1 Component Overview

The Housing component consists of expenses related to owning or
renting a dwelling. These include:
mortgage or rent payments,
utilities,
real estate taxes,
homeowner's or renter's insurance,
home maintenance, and
telephone.
At each of the three income levels, Runzheimer measured annual
housing costs under the two main housing categories: ownership and
rental.

4.2 Housing Model

4.2.1 Expenditure Research
Section 2.1.2.4 describes how Runzheimer analyzed the 1988 CES to
identify the portion of expenses attributable to each of the four
components. Runzheimer also used this survey to determine the national
average ratio of families who own, as opposed to rent, their
residences. Using the expense data from the seven most appropriate
income ranges as input into a linear regression analysis, Runzheimer
calculated own and rent weights:

Own/Rent Weightings
------------------------------------------------------------------------
Income Levels
--------------------------------------
Category Lower Middle Upper
(percent) (percent) (percent)
------------------------------------------------------------------------
Homeowner\1\..................... 37.10 46.91 62.86
Renter........................... 62.90 53.09 37.14
--------------------------------------
Totals....................... 100.00 100.00 100.00
------------------------------------------------------------------------
\1\With mortgage.

Runzheimer excluded expenditure data for homeowning families
without a mortgage because they were not typical of homeowners in the
base area or in the allowance areas with the largest concentrations of
federal employees.
The 1988 CES was also used to identify which home-maintenance items
to price and to establish the relative importance of those items.
4.2.2 Development of Housing Profiles
To compare housing costs accurately in all locations, Runzheimer
constructed a model to measure housing costs under six different
circumstances; that is, we identified six typical housing profiles and
matched these profiles to three income levels, as shown in the table
below. Runzheimer and OPM agreed that at least one criterion for the
owner profile should be the square footage of the home and at least one
criterion for the renter profile should be the number of bedrooms in
the rental unit. The profiles for homeowners and renters are:

Housing Profiles
------------------------------------------------------------------------
Income level Renter profile Owner profile
------------------------------------------------------------------------
Lower....................... 3-1-1 600\1\ sq. ft. 4-2-1 900 sq. ft.
apt.. Condo or detached
house.
Middle...................... 4-2-1 900 sq. ft. 5-3-1.5 1,300 sq.
apt.. ft. detached
house\2\
Upper....................... 4-2-2 or 5-3-2 1,100 7-3-2 1,700 sq. ft.
sq. ft. townhouse detached house.
or detached house.
------------------------------------------------------------------------
\1\Defined as ``Total rooms--Bedrooms--Baths and representative size.''
Total rooms excludes bathrooms, hallways, entrance areas, and closets
but includes bedrooms, living room, family room, kitchen, formal
dining room, and den/study. The representative size is roughly the
midpoint size for the range of housing surveyed at the income level.
\2\Row houses may be used in Northeast Washington, DC, where the
availability of single family detached homes is limited.

4.2.3 Living Community Selection
Runzheimer surveyed the same living communities and data sources
for the summer 1993 survey as it did for the 1992-1993 study with the
following exceptions:
broadened the housing survey for St. Croix, Virgin
Islands;
gathered comparable home-sale data from additional
appraisers in selected points in Maryland;
arranged for additional data sources for comparable home
sales at the upper income level in San Juan;
added a new on-line data source for comparable home-sale
prices for all the islands of Hawaii.
To gather data, our researchers contacted real estate brokers,
residential appraisers and other knowledgeable real estate
professionals in each area to obtain information on the predominant
age(s), size(s) and type(s) of housing in various communities and
housing subdivisions. When available and appropriate, Runzheimer
identified at least six communities (two at each income level) in each
allowance area. However, this goal was not achievable in some of the
smaller allowance areas. For the Washington, D.C., area, Runzheimer
selected at least nine communities (three at each income level) in
which to gather renter and homeowner data.
A table of living communities used for pricing can be found in
Appendix 6.
4.2.4 Identification and Quantification of Housing-Related Expenses
From the 1988 CES, Runzheimer identified and categorized housing-
related expense items into one of five groups:
utilities,
real estate taxes,
owners/renters insurance,
maintenance, and
telephone.
4.2.4.1 Utilities
For this study, Runzheimer classified electric, heat (oil or gas),
water and sewer as utilities. Although most utility companies had ready
access to current charges per unit of consumption and average
consumption patterns for all households, very few (if any) separated
consumption patterns by number of family members in a household or by
size/type of accommodation.
Runzheimer focused on average annual consumption experience per
household, gathering this information from utility companies serving
each allowance area and the Washington, D.C., area. Combining this
consumption data with current utility rates, Runzheimer computed
average annual utility costs for each of electric, gas or oil
(whichever Runzheimer found to be more widely used, if used at all),
water and sewer. Runzheimer then assigned this average consumption
pattern to the homeowner profile at the middle income level.
Because some utility costs vary by size of house and yard,
Runzheimer calculated a multiplier consistent with the standard home
sizes to arrive at utility rates for the other five profiles. The table
below shows the standard sizes and utility factors for each profile.
The standard sizes roughly equate to the reference size of each
profile. The formula to calculate each multiplier was:
Multiplier = 1+(.5 x (Standard square feet--1300)/1300)
The resulting utility multipliers are:

Utility Multipliers
----------------------------------------------------------------------------------------------------------------
Renter profile Owner profile
Income level ---------------------------------------------------------------------------
Square feet Multiplier Square feet Multiplier
----------------------------------------------------------------------------------------------------------------
Lower............................... 600 .73 900 .85
Middle.............................. 900 .85 1,300 1.00
Upper............................... 1,100 .92 1,700 1.15
----------------------------------------------------------------------------------------------------------------

4.2.4.2 Real Estate Taxes
For this study, Runzheimer contacted the city assessor in each
allowance area to obtain real estate tax information on the selected
living communities. (See Appendix 6 for a listing of the communities.)
Real estate tax formulas were obtained for most living communities;
however, Runzheimer's researchers found that in San Juan, PR, the
formulas do not always translate into the actual taxes paid by typical
property owners.
Therefore, in San Juan, Runzheimer undertook a comparative study of
current actual taxes paid by homeowners and their current home values.
Runzheimer then developed an average ratio for taxes paid to current
home values. This ratio was applied to the average home value in that
community to obtain an average real estate tax amount for the desired
homeowner profile.
4.2.4.3 Owners/Renters Insurance
As it did for previous surveys, Runzheimer undertook to gather
insurance-rate information for the allowance areas for both renter and
owner profiles. These rates represent coverage for structure and
contents for homeowners but contents only for renters.
Previous research conducted by Runzheimer, at the request of OPM,
found that insurance coverage for disasters, such as floods and
earthquakes, is not commonly purchased by allowance area residents.
Consequently, Runzheimer, with the concurrence of OPM, does not
consider these additional riders. (See Report to OPM on Living Costs in
Selected NonForeign Areas and in the Washington-D.C., Area, June 1992
at 57 FR 58556). Runzheimer notes, however, that OPM is reviewing the
results of the Federal Employee Housing and Living Patterns Survey as
they apply to this issue.
4.2.4.4 Maintenance
Many factors were involved in measuring the cost of maintaining a
home, including area climate, architecture and building materials, and
the cost of maintenance materials and labor. As it did for the previous
survey, Runzheimer priced such household maintenance commodities as
fire extinguisher, bathroom caulking and kitchen faucet. Pest control
service was priced in each allowance area surveyed.
Runzheimer developed maintenance costs based on the cost of
maintenance materials and labor rates in each area. Runzheimer's
approach to maintenance was the same as the approach to goods and
services, as explained below.
Runzheimer used expenditure data from the 1988 CES to identify the
national average home-maintenance expense, the maintenance items to
survey, and the appropriate item weighting. Because most, if not all,
maintenance items were included in rent, maintenance costs were not
added in the three renter profiles.
To compute home-maintenance cost differences between each allowance
area and the Washington, DC, area for the homeowner profiles,
Runzheimer obtained prices for selected building materials and labor
rates for maintenance work. For each area, Runzheimer computed the
relative cost (i.e., an index) for each maintenance item compared to
the cost of that item in the DC area. As with Goods & Services, the
results of the nationwide CES were used to weight these maintenance
indexes into an overall index for each area.
To combine maintenance indexes with the other homeowner costs,
which were expressed in dollar amounts, Runzheimer converted the
indexes to dollars. To do this, Runzheimer multiplied the maintenance
cost index for each area by the CES nationwide average maintenance cost
and assigned that cost to the middle-income homeowner profile.
Logically, maintenance costs for larger homes would be greater than
costs for middle-sized homes, while costs for smaller homes would be
less. Therefore, in this study, Runzheimer applied the same homeowner
multipliers used in the utilities model for the lower and upper income
profiles (.85 and 1.15 respectively) to recognize differences in
maintenance costs due to house size.
4.2.4.5 Telephone
Telephone expenses consisted of local service charges, possible
additional charges for local calls, and charges for long distance
calls. To measure estimated expenses for local service and local calls,
where available, Runzheimer surveyed the cost of touch-tone service
with unlimited calling.
To estimate long distance charges in all areas, Runzheimer surveyed
the cost of three ten-minute direct dial calls per month to large U.S.
mainland cities (i.e., Los Angeles, Chicago, and New York City).
Runzheimer measured the price of a call placed in the survey area at
the time of day necessary to be received in the respective city at 8:00
p.m. local time. In many areas, this resulted in pricing a combination
of daytime and evening-rate calls.
4.3 Housing Data Collection Procedures
As was done in previous years, Runzheimer collected housing
information from a variety of sources. Also as in previous years,
Runzheimer research personnel traveled to profiled communities in the
allowance areas to observe first hand the comparability of homes in one
area versus another and the appropriateness of individual housing units
for the profiled income level.
Last year, OPM modified the contract to require several additional
steps to increase the quantity and quality of housing data collected.
These extra steps were incorporated again this year and included
purchasing data from additional real estate listing services, making
greater use of assessor's records when other data were not available,
and contacting more real estate professionals to obtain additional home
sales, rental data, and market trend information.
4.3.1 Homeowner Data Collection
In the homeowner data-gathering phase, Runzheimer obtained sale
prices of homes (called ``comparable sales'') in the area that matched
the housing profiles. In the communities that were identified (see
section 4.2.3), Runzheimer tried to obtain all the comparable sales
during the 6-month period prior to the date of the survey. For the
surveys covered in this report, the home sales pricing period was
January 1993 through July 1993.
As was done last year, Runzheimer contacted knowledgeable and
helpful real estate professionals in each location and/or used real
estate sales data and listing services. The amount of data obtained
depended on the number of home sales in the community and the
availability of square footage and other information. This in turn
depended on the size of the community, the economic conditions, the
quality and quantity of the realty data available, and the willingness
and ability of local realty professionals and assessor offices to
provide data. If the comparable sales data obtained from the first data
sources were insufficient, Runzheimer contacted additional data sources
in the area to attempt to secure more sales data, if practical.
4.3.2 Renter Data Collection
In some cases, the same Realtors and brokers who assisted in our
profiling phase were very active in the rental markets as well. When
this occurred, Runzheimer obtained current rental rates and fees for
our profiled apartments, townhouses, and houses from these sources.
Runzheimer also contacted rental management firms that operate
apartment complexes matching the profile specifications. In large
metropolitan areas, such as the Washington, D.C., area where rental
complexes abound, our housing analysts conducted telephone surveys to
obtain current rental information.
OPM modified the contract last year to expand the level of effort
that Runzheimer expended in the collection of housing data. The result
both last year and this year was a marked increase in the quantity of
housing data collected for both rentals and sales.
Rental data were obtained from a variety of sources, e.g., brokers,
property managers, newspaper advertisements, and others. Analyses of
these data indicated that there are two separate rental markets--a
broker market and a non-broker market. Rental rates and estimates
provided by brokers generally exceed those obtained from other sources.
In each area, the quantity of data obtained from either source-type
varies significantly. Therefore, analyzing all of the rental data (both
broker and non-broker) together for an area and income level is
undesirable. Because OPM has no information on how federal employees
who rent generally secure their lodgings, Runzheimer applied equal
weights to the broker and non-broker data to compute the overall
average rental rate for the area and income level. (See Appendix 9B.)

4.4 Housing Analysis

4.4.1 Homeowner Data Analysis
One of the most important factors relating to the price of a home
is the number of square feet of living space. For each income profile
in each allowance area and the Washington, D.C., area, Runzheimer
computed that average price per square foot for the comparables. Except
as noted below, Runzheimer used this value times the reference square
footage for the profile to determine the average home value for the
profile.
Runzheimer experienced difficulties in obtaining housing data in
some areas. For example, despite several efforts to obtain more
detailed information from various sources, Runzheimer was able to
obtain very few comparable sales with square footage information for
home sales in Mayaguez. Also, as noted earlier in this report, OPM
elected to combine the St. Croix and St Thomas data to address concerns
about the number of home-sale observations obtained in the Virgin
Islands.
4.4.1.1 Data Trimming
Based on experience from the previous home-pricing surveys, last
year OPM modified the living-cost model as it applied to the analysis
of housing data. The modifications allowed the use of housing costs
trend data as well as current housing costs in the analysis of owner
and renter living costs. These analyses are consistent with section
591.205(b)(3) of title 5, Code of Federal Regulations, and result in
improved housing data results.
One of the modifications involves ``trimming'' the observations.
Runzheimer and OPM noted that a relatively few extreme values (values
that were either exceptionally low or high cost) could have a
significant influence on the average housing costs observed at an
income level within an area. Including these extreme values had the
potential to cause results to vary erratically from one year to the
next.
In situations such as this, statisticians frequently use the median
or trim the data in some manner to reduce its volatility. The use of
the median home value was not desirable because some areas had
relatively sparse data at one or more income levels. This could make
the median unstable from one year to the next.
Last year Runzheimer recommended, and OPM agreed, that the
observations be ranked from low to high on the basis of the cost per
square foot and that the top and bottom 20% of the observations be
``trimmed'' (i.e., eliminated) from the data before averages or trends
were calculated. (Data were not trimmed if there were four or fewer
observations.) These procedures reduce the influence of home sales
anomalies and make survey results more stable from one year to the
next.
4.4.1.2 Special Considerations
The new procedures also involved analyzing data in a more thorough
and integrated manner. The procedures required analyzing the current
housing survey data, analyzing the trends observed when these data were
compared with the previous survey's data, and comparing these trends
with the views obtained from real estate professionals in the area. How
and which data were used depended on the quality and quantity of data
collected and how the trends observed agreed among income levels and
with the views of local real estate professionals. These procedures are
discussed below.
Runzheimer sought to gather all of the appropriate comparable sales
data available in each area. As a minimum, Runzheimer sought to obtain
10 Realtor sales per community per income level or 20 per income level
per area. In many areas, the sales data exceeded the minimum.
If the minimum number could not be obtained or if highly divergent
trend data were observed among income levels in the area or as compared
with the views of local real estate professionals, additional analyses
were performed. These analyses were:
1. If the current data were significantly better than the previous
data (e.g., greater in quantity or more consistent), the current data
were used to the extent practical.
2. If at least three observations at each income level were
available, and the previous data were better than the current data or
the previous and current data were of equivalent quality, the change
(i.e., trend) in the average price was used to update the previous
data. This was done using one of the following procedures, depending on
the situation:
a. If data problems occurred at one income level only, the average
rate of change at the other two income levels was used to adjust the
previous prices for the affected income level.
b. If data problems occurred in two income levels, the rate of
change observed at the non-problem income level was used to adjust the
previous prices for the two affected income levels.
c. If data problems occurred at all income levels, the average rate
of change observed at all three income levels was used to adjust the
previous prices at all income levels.
3. If fewer than three observations were available at any income
and/or the data quality was questionable at all levels, all three
levels of current and previous survey data were merged. These data were
then analyzed by applying the procedures described in section 4.4.1.1
above to each set of merged data, and the average cost per square foot
was computed for each set of data and compared to estimate the overall
change in the area. This overall change was applied to the previous
average costs by income level to determine the current costs at each
income level for the area.
The areas for which these procedures were applied and the
calculations used are found in Appendix 9A.
4.4.2 Rental Data Analysis
Runzheimer assigned each rental quote data point to a single income
level, based on these criteria:
Assign one bedroom apartments to the lower income level.
Assign two bedroom apartments to the middle income level.
Assign townhouses and detached houses with a minimum of
two bedrooms to the upper income level.
4.4.2.1 Data Trimming and Special Analyses
In the analysis of rental data, Runzheimer applied the same
procedures used to trim the home sales data (see section 4.4.1.1)
except that data were ranked on the basis of monthly rental rates, not
cost per square foot. Also, as with home sales analyses, special
analyses were applied to rental data when the data where sparse or
highly divergent trends were observed among income levels. These
analyses were the same as those applied to home sales data (see section
4.4.1.2) except that if data were merged, overall estimates were based
on monthly rental rates, not the cost per square foot. (See Appendix 9B
for these analyses.)
4.4.3 Analysis of Housing-Related Expenses
Because section 4.2.4 covers the identification and quantification
of housing-related expenses, these topics are not repeated here.
However, it should be noted that Runzheimer incorporated home sale
prices from this study into the calculations of real estate taxes and
homeowner's insurance, which depend upon the value of the home.
4.5 Housing Survey Results
In the above sections, Runzheimer describes how it measured the
costs for maintenance, insurance, utilities, real estate taxes, rents,
and homeowner mortgages. Appendix 7 shows the cost of each of these
items, for renters and homeowners separately, in each allowance area
and in the Washington, D.C., area. For Hawaii County, Hawaii; Puerto
Rico; and the Virgin Islands, these costs are shown separately for each
of the survey areas within the allowance area.
Appendix 8 compares the total cost of these items in each allowance
area with the total cost of the same items in the Washington, D.C.,
area. Again, there are separate comparisons for renters and homeowners.
The final housing-cost comparisons take the form of indexes that
are used in Appendix 14 to derive the total, overall index for owners
and renters. (Refer to section 2.2 for a discussion of the general
formulae and how the component indexes are combined.)

5. Transportation

5.1 Component Overview
The Transportation component consisted of expenses related to
private and public transportation. The private transportation category
contained expenses related to owning and operating a vehicle in each
area. The public transportation category focused on the cost of air
fares from each location to a common point within the contiguous 48
states.
As in previous surveys, Runzheimer used national average
expenditure data to combine the private and public transportation
relative cost differences between each allowance area and the
Washington, D.C., area to arrive at a total Transportation component
index.
5.2 Private Transportation Methodology
Runzheimer determined that an accurate and reasonable approach to
measure transportation costs was to select and analyze three commonly
driven vehicles (a domestic auto, an import auto and a utility vehicle)
in all areas.
New vehicles were the basis for developing the transportation-cost
calculations. Although Runzheimer could have developed costs from the
premise that ``identical'' used vehicle would be purchased from auto
dealers in each location, Runzheimer believed that costing new vehicles
reduced the potential for inconsistencies due to value judgments
concerning used vehicles.
5.2.1 Vehicle Selection and Pricing
As mentioned above, Runzheimer selected and priced a domestic auto,
an import auto, and a utility vehicle as the basic vehicle types to
cost in all locations. We based our selection of these vehicle types on
their popularity in the United States as demonstrated by owner
registration data.
To select a specific make and model within each vehicle type,
Runzheimer identified the top-selling models in each car class. For
these models, Runzheimer's research associates collected new vehicle
prices.
At each auto dealership in the sample, Runzheimer recorded the
suggested retail prices of the three vehicles plus any additional
charges, such as shipping, excise tax, dealer prep, and additional
dealer markup. Runzheimer used the suggested retail prices (not
negotiated prices) in the analysis. Runzheimer also included
documentation fees as part of the new-vehicle costs in Hawaii.
Contacted dealerships explain that a documentation fee is charged on a
new-car purchase to cover paperwork costs. Runzheimer did not include a
fee for the Washington, D.C., area and other tropical areas because
dealers in those areas do not typically charge a documentation fee.
The three vehicles selected for analysis were:

Domestic Vehicle--Ford Taurus GL 4-door sedan 3.0L 6cyl
Utility Vehicle--Chevrolet S10 Blazer 4X4 2 door 4.3L 6 cyl
Import Vehicle--Honda Civic DX 4-door sedan 1.5L 4 cyl

Runzheimer priced 1993 models in this survey. All vehicles were
equipped with standard options, such as automatic transmission, AM/FM
stereo radio and air conditioning.
Car dealers in the Washington, D.C., area do not recommend vehicle
rustproofing. However, it is suggested or recommended in allowance
areas. Therefore, we include rustproofing as an add-on in all allowance
areas, but not in the Washington, D.C., area.
5.2.2 Vehicle Trade Cycle
Calculating the cost to own and operate a vehicle requires that two
important factors be determined: miles driven and time period of
ownership. In the automobile industry, these two factors are known
collectively as a vehicle's ``trade cycle.'' The trade cycle is stated
as a length of time either in months or years, and the total number of
miles driven in that time period (e.g., four-year, 60,000-mile trade
cycle). This information is required to compute annual costs related to
fuel, oil, tires, maintenance and depreciation.
Conforming with previous living-cost reports, Runzheimer used a
four-year 60,000-mile trade cycle in all areas based upon the following
information:
The Internal Revenue Service has used this trade cycle for
many years to compute the allowable cents-per-mile reimbursement rate
for persons who drive their personal vehicle for business purposes.
The four-year time period coincides with the typical
length of a vehicle loan.
U.S. Department of Energy statistics for 1988 show that
the U.S. average for number of vehicles miles driven was: 18,595 per
household and 10,246 miles per vehicle.
Runzheimer has been unable to find conclusive statistics on average
annual miles driven per vehicle in any allowance area. In the past,
Runzheimer contacted car dealers to obtain their observations on
average odometer mileage on trade-in vehicles.
From the opinions gathered, we concluded that, in most cases, the
average annual miles driven in allowance locations appeared to be less
than or equal to 15,000. In the Washington, D.C., area, the opinions of
those contacted indicated an average annual mileage of 15,000 or more.
Therefore, without definitive statistics to prove otherwise, Runzheimer
set a standard used in all reports to date of 15,000 miles per year,
which results in a four-year, 60,000-mile trade cycle.
5.2.3 Fuel Performance and Type
To establish average fuel-performance ratings, Runzheimer selected
the ``city driving'' figures published by the Environmental Protection
Agency (EPA). Runzheimer chose the ``city'' instead of ``highway''
figures because all locations contained considerable stop-and-go
driving conditions. All vehicles included in this study used regular
unleaded fuel. Runzheimer obtained self-service cash prices at branded
stations only, and substituted full-service when self-service was not
available or only non-branded stations were available.
As in its second report to OPM, Runzheimer has included in its
analysis a number of fuel-performance factors; specifically,
temperature, road surface, and gradient. Based on our research of these
three factors, Runzheimer analysts developed fuel-performance
adjustment percentages in each allowance area.
5.2.3.1 Impact of Temperature upon Fuel Performance
Runzheimer consulted two published sources to develop its
adjustment percentages for this fuel-efficiency factor: Passenger Car
Fuel Economy: EPA and Road and The Weather Almanac (Ruffner & Blair).
Miles-per-gallon performance varies by ambient temperature. The lower
the temperature, the fewer miles-per-gallon achieved and vice versa. In
the EPA study, the temperature at which no adjustments to fuel
performance occur is 77 deg. F. Below that temperature, miles-per-
gallon achieved drops; above 77 deg., miles-per-gallon achieved
improves. To measure the effect temperature has on miles-per-gallon for
each allowance area, Runzheimer research average monthly temperatures
as reported in The Weather Almanac.
In each location and for each month, Runzheimer assigned the
appropriate shortfall factor from the EPA study based on the average
monthly temperature for each given location. After assigning factors to
each month, Runzheimer averaged the twelve factors for each location.
The results of these calculations are shown in section 5.2.3.4.
5.2.3.2 Impact of Road Surface upon Fuel Performance
For its analysis, Runzheimer assumed that federally controlled
roadways are typically composed of concrete and/or high-load asphalt
and that locally controlled roadways are typically composed of low-load
asphalt. EPA's research indicates that cars are generally more fuel-
efficient on the firmer, high-load surfaces than on the softer, low-
load surfaces.
Although traffic patterns and road usage certainly vary among
areas, Runzheimer could find no relevant studies of these issues.
Therefore, Runzheimer assumed that federally controlled roadways
generally support twice the traffic of or are used at least twice as
much as locally controlled roadways.
In each allowance area, Runzheimer researched the total mileage
falling into either the federal or local categories. For example,
Hawaii contains 1,456 miles of federally controlled roads and 2,606
miles of locally controlled roads. The usage assumption allowed
Runzheimer to increase federal road mileage by a factor of two.
Runzheimer applied the average low-load asphalt factor (which
reflects dry, wet, and snowy conditions) to the local mileage
percentage and the average concrete and/or high-load asphalt factor to
the federal mileage percentage to create a weighted average factor for
each area. The weighted factor for the allowance areas surveyed was
0.98. The Washington, D.C., area was assigned a factor of 1.00 on the
premise that the vast majority of traffic in that area travels on dry,
high-load surfaces. (See section 5.2.3.4 for application of this factor
in estimating overall MPG.)
5.2.3.3 Impact of Gradient upon Fuel Performance
Runzheimer consulted EPA's Passenger Car Fuel Economy: EPA and Road
to determine the effect of local topography (i.e. gradient) upon fuel
efficiency. EPA provides mileage factors based upon various gradients
ranging from less than 0.5% (essentially flat) to greater than 6%
(steep).
Runzheimer reviewed the topographic features of each area and found
a wide range of road conditions. However, Runzheimer was unable to find
information on the types of terrain drivers typically encounter in each
area or the number of miles drivers travel in each type of terrain.
Lacking such information, Runzheimer assumed that drivers in the
allowance areas generally travel roads having approximately the same
gradients that are found on average in the United States. Applying the
information from EPA's research, Runzheimer computed a fuel-performance
factor of 0.981 for this type of driving. This factor was assigned to
each allowance area. Runzheimer assigned a factor of 1.00 to the
Washington, D.C., area on the premise that the vast majority of traffic
in that area travels on major freeways and highways that are relatively
flat. (See section 5.2.3.4 for application of this factor in estimating
overall fuel efficiency.)
5.2.3.4 Overall Impact upon Fuel Performance
Runzheimer applied the results of the analyses described above to
``localize'' or make geographically sensitive adjustments to the EPA
average ratings and establish reasonable fuel-performance ratings for
each allowance area.
In the table below, the factor 1.00 means that no adjustment to EPA
fuel performance is appropriate. A factor of less than 1.00 means that
the estimated gasoline mileage in the area is less than the EPA
average. For example, the total adjustment factor for Guam is 0.95.
This means that the estimated gasoline mileage in Guam is 95% of the
EPA estimated average. Note that the adjustment factor for the
Washington, D.C., area (0.94) indicates that average gasoline mileage
in that area is below the EPA estimate also.

Summary of Fuel-Performance Adjustments
------------------------------------------------------------------------
Road
Location Temperature Surface Gradient Total
------------------------------------------------------------------------
Hawaii....................... 0.99 0.98 0.98 0.95
Guam......................... .99 .98 .98 .95
Puerto Rico.................. 1.01 .98 .98 .97
Virgin Islands............... 1.01 .98 .98 .97
Washington, D.C.............. .94 1.00 1.00 .94
------------------------------------------------------------------------
Note: These adjustments compound. That is, the Total adjustment is the
result of multiplying the three individual factors together for each
location/area.

5.2.4 Vehicle Maintenance
With OPM's concurrence, Runzheimer selected the five most common
maintenance service/repair jobs performed on vehicles as the basis for
vehicle maintenance analysis:
tune-up,
oil change,
automatic transmission fluid change,
flush/fill coolant, and
muffler installation.
Automobile manufacturers' recommended maintenance schedules were
used to determine the frequency of performing each of these maintenance
jobs. Maintenance schedules vary, depending on the driving conditions
typically encountered. Consistent with the assumptions used for fuel
economy and tire mileage, Runzheimer assumed that driving conditions in
the allowance areas were generally severe and used the maintenance
schedules that reflected that kind of driving. For the D.C. area,
Runzheimer assumed that driving conditions were normal and used the
maintenance schedules that reflected that kind of driving.
The recommended frequency of performing each of these jobs was
combined with the prices charged by local dealers and service stations
to compute an estimated annual maintenance expense.
For this year's survey, Runzheimer collected specific parts costs
and hourly labor costs in each location, three prices per item, per
location. Runzheimer used Chilton's Labor Guide and Parts Manual to
determine service times for each maintenance procedure, and used only
one service time per maintenance procedure. This will result in more
uniform prices within each location.
5.2.5 Tires
Research previously conducted by Runzheimer for OPM (see the June
1992 report) revealed that various factors (e.g., road quality/state of
repair, road composition) caused tread life (the average number of
miles a tire is expected to last) to be less in allowance areas than in
the Washington, D.C. area. Based on these findings, Runzheimer based
tire expenses on a 40,000-mile tread-life in allowance areas, and a
55,000-mile tread-life in the Washington, D.C., area.
5.2.6 License and Registration Fees, and Miscellaneous Tax
Runzheimer obtained information regarding appropriate license and
registration fees, and miscellaneous taxes (i.e., personal property tax
and motor vehicle registration tax) from each area. One-time fees and
miscellaneous taxes were divided equally over each vehicle's four-year
trade cycle. Sales and excise taxes were included in the purchase price
of each vehicle (see section 5.2.7.). Ongoing fees and taxes were
included as part of the annual costs.
5.2.7 Depreciation
From Runzheimer's experience, the single largest annual expense
related to owning and operating newer vehicles is vehicle depreciation,
the lost value of the vehicle as it ages and is driven. To calculate
average annual depreciation, Runzheimer divides the difference between
the purchase price and the residual value by the number of years the
vehicle is owned.
In the depreciation equation, Runzheimer used suggested retail
prices, plus any additional charges, such as shipping, excise tax,
dealer prep, and additional dealer markup. (Runzheimer did not believe
that negotiated prices could be collected on an equitable basis.) As
discussed earlier, the trade cycle was determined to be four years,
60,000 miles. Runzheimer research indicated that residual values were
the same in all areas. This research effort is explained below.
Runzheimer is aware that several firms and associations track and
publish weekly or monthly used-car and used-truck wholesale auction
prices. Some firms even publish projections of the future value of
today's new vehicles. Most publications provide several residual values
for each vehicle, depending on its condition at the time of trade-in
(e.g., clean, average, rough). Several common publications of this type
are Black Book, Kelley Blue Book, Automotive Market Report, and NADA
National Automobile Dealers Association). Unfortunately, these sources
only track prices for vehicles sold in the contiguous 48 states and
then publish broad-based average residual values for each vehicle.
To get specific information from sources knowledgeable about the
used vehicle markets in allowance areas, Runzheimer contacted auto
dealers and financial institutions in these areas. Most of the sources
with whom Runzheimer spoke said that they used the above-mentioned
publications as guides, just as dealers and financial institutions
across the United States used them.
Runzheimer found no conclusive evidence that used vehicles in
allowance areas were (on average) worth more or less than used vehicles
in the Washington, D.C., area. Therefore, we reported the same used
vehicle prices in all areas. An appropriate and logical source for
these values was the April 1993 issue of Black Book Official Finance/
Lease Guide for 1993 vehicles.
It should be noted for clarification that identical residual values
did not translate into identical depreciation amounts in all locations.
Depreciation amounts were higher in allowance areas than in the
Washington, D.C., area because new vehicle prices in all allowance
areas were higher. For example, new vehicle prices in Puerto Rico
averaged 50% more than Washington, D.C., prices.
5.2.8 Finance Expense
Runzheimer included the average annual cost of financing a vehicle
in the total cost of private transportation. Runzheimer surveyed
automobile dealerships in Puerto Rico and banks in all other areas for
their auto-loan interest rates, using a 48-month loan length with 80%
financing as the basis in all locations.
5.2.9 Vehicle Insurance
Runzheimer measured the cost of auto insurance in each location. To
determine the type of coverage to price, Runzheimer contacted insurance
agents in each area to obtain information on the typical policy. Listed
below are the most common coverages, limits, and deductibles for the
surveyed living-cost areas.

Bodily Injury--$100,000/$300,000
Property Damage--$50,000
Medical--$5,000
Uninsured Motorist--$100,000/$300,000
Comprehensive--$100 Deductible
Collision--$250 Deductible

Runzheimer found that insurance companies in Guam, Puerto Rico, and
the Virgin Islands provide slightly different limits and deductibles
than those listed above; therefore, Runzheimer and OPM agreed to
incorporate the premiums associated with these different limits and
deductibles into the indexes.
To do this, Runzheimer surveyed in the D.C. area the price of
insurance policies that were equivalent to the policy offered in the
allowance area. From this comparison, Runzheimer computed an index that
was applied to the price of the typical policy surveyed in the
Washington, DC, area. By applying this factor to the average price in
the D.C. area, Runzheimer was able to estimate the cost of equivalent
coverage in the allowance area. (See Appendix 10B.)
In all areas, Runzheimer attempted to identify the most ``popular''
automobile insurance companies by analyzing market-share reports
compiled by an industry rating bureau. The policy described above was
then priced again this year for each location. Two or three price
quotes were obtained for each area, totalled for each area, and
averaged together to produce the final number for this component in
each allowance area.
5.3 Public Transportation Methodology
As was done last year, Runzheimer surveyed the cost of air fares as
they relate to recreational travel. Runzheimer priced the lowest
available round-trip air fare from each allowance area and the
Washington, D.C., area to Los Angeles, California. Los Angeles was
selected because it is a common point approximately equidistant from
most of the allowance areas and the Washington, D.C., area. The cost of
the trip from each allowance area to Los Angeles was compared with the
cost of the trip from the Washington, D.C., area to Los Angeles to
compute the public transportation category indexes. (See Appendix 11.)
5.4 Transportation Survey Results
Runzheimer measured the costs for fuel, maintenance and oil, tires,
licensing, taxes, depreciation, finance, and insurance for three types
of automobiles in each allowance area and in the Washington, D.C., area
to determine typical private transportation costs. Appendix 10A shows
the cost of each of these items in each area. As with the housing
costs, private transportation costs for Hawaii County, Puerto Rico, and
the Virgin Islands, are shown separately for each of the survey areas
within these allowance areas. These data are combined to produce
composite costs for each allowance area using the respective federal
employment distributions. (See sections 2.5 and 2.6 for a discussion of
employment weighting in these two areas.)
Appendix 11 compares the total cost of the private transportation
items for each vehicle in each allowance area with the total cost of
the same items in the Washington, D.C., area. Appendix 11 also shows
how the private and public transportation indexes were combined using
expenditure weights derived from the CES data to produce final
transportation indexes.
The final transportation indexes are used in Appendix 14 to derive
the total overall index. (Refer to section 2.2 for a discussion of the
general formulae and how the component indexes are combined.)

6. Miscellaneous Expenses

6.1 Component Overview
The Miscellaneous Expense component consists primarily of four
unrelated groups of expenses:
Medical care,
Contributions (including gifts to non-family members),
Personal insurance, and
Savings and investments (including pensions).
Runzheimer believes that certain miscellaneous expense items should
not affect living-cost differences between locations. For example,
Runzheimer considers charitable contributions a personal choice, so we
include this expenditure as a constant amount in all locations. Based
on research into all of the expenses of this component, Runzheimer also
regards expenses related to personal insurance, savings and
investments, and pensions as constants, for reasons discussed in
section 6.2.2.
To measure the miscellaneous expenses, Runzheimer constructed a
pricing methodology similar to the one used in the Goods & Services
component. Runzheimer selected representative items for medical care,
priced them in all areas, and then computed a Miscellaneous Expense
component index based on the relative importance of costed items/
categories held constant.
6.2 Miscellaneous Expense Model
6.2.1 Expenditure Research
From the 1988 CES, Runzheimer tabulated the miscellaneous expense
data into logical expense groupings and then determined the appropriate
item weighting. The table on the following page lists the categories
that Runzheimer selected to price and their weights:

Miscellaneous Expense Categories & Weights
------------------------------------------------------------------------
Income level
--------------------------------
Categories Lower Middle
(percent) (percent) Upper
(percent)
------------------------------------------------------------------------
Medical Care........................... 43.41 31.56 22.40
Contributions (including gifts)\1\..... 12.38 14.90 16.85
Personal Insurance & Pensions\1\....... 44.21 53.54 60.75
--------------------------------
Totals........................... 100.00 100.00 100.00
------------------------------------------------------------------------
\1\Held constant.

6.2.2 Miscellaneous Expense Methodology
As stated in section 2.2, Runzheimer used the Laspeyres indexing
methodology to compute the Miscellaneous Expense component index. For
groups of items held constant, the model assumed a price ratio between
the allowance area and the Washington, D.C., area equal to 100.00%.
Runzheimer defined personal insurance and pensions as the portion
of a family's budget that was targeted for long-term financial
security. This is consistent with the definitions used by the CES, the
results of which are used as weights in the COLA model. In the CES,
money stored in a savings account or investment vehicle for future
expenditures (of goods and services, housing, or transportation) is
accounted for in the other component weightings.
In section 6.1, we noted that expenses related to personal
insurance were held constant for all locations. This was based on
information received from life insurance companies and OPM officials.
The life insurance companies contacted indicated that policies written
(and premiums charged) to persons within the United States and its
territories did not vary due to location. Runzheimer's research and
discussions with OPM officials also indicated that, in general, federal
employees in all areas received similar or identical benefits
packages--variations are generally due to personal preference.
Therefore, Runzheimer believed, and OPM concurred, that holding these
types of expenses constant was appropriate.
6.3 Miscellaneous Expense Data Collection Procedures
Medical care items were surveyed consistent with the approach used
in the Goods & Services component. For quality-control purposes,
Runzheimer used its in-house research staff to conduct much of this
survey.
The following medical-care items were priced in each allowance area
and in the Washington, D.C., area:
Nonprescription pain reliever.
Prescription drugs.
Vision check.
Dental service.
Doctor visit.
Hospital room.
Health insurance.
Runzheimer computed a Medical Care subcategory price index for each
item in each allowance area by comparing each local average price with
the Washington, D.C., area average prices. These indexes were combined
using weights derived from the CES to compute a Medical Care
subcategory index for each allowance area.
6.4 Miscellaneous Expense Survey Results
Appendix 12 contains the results of Runzheimer's data collection
and index calculations. As the appendix shows, the relative costs of
the majority of the items in the Miscellaneous Component are based on
surveyed prices. Therefore, the Miscellaneous Component index reflects
living-cost differences among areas. The cost of only two items--life
insurance/pensions and contributions--does not differ among areas.
Although these two items together have a significant weight, one should
keep in mind that the Miscellaneous Component has the smallest weight
of the four components.
Section 2.2 describes how the Miscellaneous Expense component
indexes are combined with the other component indexes to derive the
final index for each area.

7. Final Results

7.1 Total Comparative Cost Indexes
The total comparative cost indexes appear below. Appendix 14 shows
how each index was derived from the component indexes.

Final Cost Comparison Indexes
------------------------------------------------------------------------
Commissary
Allowance area Local and
pricing exchange
------------------------------------------------------------------------
City & Cnty of Honolulu, Hawaii................... 122.90 120.26
Hawaii Cnty, Hawaii............................... 109.63 NA
Kauai Cnty, Hawaii................................ 119.27 NA
Maui Cnty, Hawaii................................. 119.32 NA
Guam, CNMI\1\..................................... 122.25 120.81
Puerto Rico....................................... 103.00 102.17
U.S. Virgin Islands............................... 117.81 NA
------------------------------------------------------------------------
\1\Commonwealth of the Northern Mariana Islands.
NA=Not Applicable.

7.2 General Comments
Runzheimer's primary goal throughout its work on each study has
been to bring fairness and accuracy to the results. The scope of this
multi-year engagement has become more comprehensive by virtue of
special research projects, seasonal pricings, expanded marketbasket
pricings and other efforts. Runzheimer believes that living-cost
research is a dynamic process, not a static one, and that fresh
research and analysis will enhance further the quality of the survey
and the findings. Moreover, we believe that planned, ongoing
interaction with OPM will aid the process and improve accuracy.
7.3 Recommendations
As noted earlier in this report, Runzheimer and OPM are researching
the issue of including income taxes in the living-cost surveys and
analyses. We believe that the research will show that income taxes
represent a significant portion of living expenses--a portion that
varies from one area to the next.
As also noted in the report, Runzheimer recognizes that it applied
the same salary levels and CES data this year as it did in the 1990
surveys. We commend OPM for introducing new Federal employment weights
and urge OPM to continue with its plans to introduce gradually new CES
data and salary levels in future surveys.

Appendix 1.--Consumer Expenditure Survey (CES)
[By Income Before Taxes: Average annual expenditures and characteristics of all consumer units, Consumer
Expenditure Survey 1988, Feb. 13, 1990]
----------------------------------------------------------------------------------------------------------------
June 7, 1990
------------------------------------------------------------------------------------------
Item Total
complete $10,000 to $15,000 to $20,000 to $30,000 to $40,000 to $50,000 and
reporting $14,999 $19,999 $29,999 $39,999 $49,999 over
----------------------------------------------------------------------------------------------------------------
Number of consumer
units (in thousands) 81354 9433 8219 14586 10901 7198 12209
Number of sample
interviews.......... 30900 3500 3107 5496 4119 2849 4983
Consumer unit
characteristics:
Income before
taxes\1\.......... $28540 $12320 $17373 $24591 $34375 $44331 $74234
Income after
taxes\1\.......... 26149 11892 16345 22963 31660 40100 66345
Average number of
persons in
consumer unit..... 2.6 2.2 2.5 2.7 2.9 3.2 3.1
Age of reference
person............ 46.9 50.1 46.5 44.7 43.2 42.3 45.3
Average number in
consumer unit:
Earners.......... 1.4 0.9 1.2 1.5 1.8 2.0 2.1
Vehicles......... 2.0 1.4 1.9 2.2 2.6 2.7 3.1
Children under 18 0.7 0.6 0.7 0.7 0.9 1.0 0.8
Persons 65 and
over............ 0.3 0.5 0.4 0.3 0.2 0.1 0.1
Percent
distribution:
Male........... 66 57 64 71 78 82 87
Female......... 34 43 36 29 22 18 13
Homeowner with
mortgage...... 38 15 26 36 52 64 76
Homeowner
without
mortgage...... 24 32 26 25 18 14 14
Renter......... 39 53 47 39 30 21 11
Black.......... 11 12 10 10 5 6 4
White and other 89 88 90 90 95 94 96
Elementary (1-
8)............ 11 17 12 8 5 2 3
High school (9-
12)........... 44 51 54 48 42 40 24
College........ 44 31 34 44 53 58 73
Never attended
and other..... 1 1 1 0 0 0 0
At least one
vehicle owned... 86 84 91 95 96 97 97
Average annual
expenditures........ 26389.07 16788.64 19558.35 24896.36 31659.60 37562.00 52320.19
Food............... 3804.39 2777.33 3194.53 3765.02 4587.49 5281.61 6296.11
Food at home..... 2176.94 1809.23 1954.49 2174.01 2556.74 2906.55 3109.86
*Cereals and
bakery
products...... 317.03 266.20 274.62 320.55 375.38 417.06 450.19
*Cereals and
cereal
products.... 111.15 101.45 100.46 111.31 134.59 145.71 138.66
*Flour....... 4.83 6.43 4.59 4.99 5.06 4.15 4.17
*Prepared
flour mixes. 9.88 9.30 9.21 10.32 11.92 14.72 12.18
*Ready-to-eat
and cooked
cereals..... 73.49 65.38 65.31 72.80 89.56 98.06 92.85
*Rice........ 7.98 8.00 6.06 7.95 9.66 9.48 10.14
*Pasta,
cornmeal and
other
cereals..... 14.97 12.33 15.29 15.24 18.39 19.30 19.34
*Bakery
products...... 205.88 164.75 174.16 209.23 240.80 271.35 311.53
*Bread....... 65.72 58.48 61.24 68.58 72.19 78.50 86.03
*White
bread..... 35.48 32.79 33.61 38.12 39.98 39.46 42.54
*Bread,
other than
white..... 30.24 25.69 27.63 30.46 32.21 39.05 43.49
*Crackers and
cookies..... 51.76 41.43 42.92 53.39 60.40 75.75 77.18
*Cookies... 32.19 24.30 27.98 33.01 35.93 47.99 49.96
*Crackers.. 19.57 17.13 14.94 20.38 24.46 27.76 27.21
*Frozen and
refrigerated
bakery
products.... 13.55 10.10 10.54 13.12 15.29 17.64 24.89
*Other bakery
products.... 74.84 54.74 59.46 74.14 92.92 99.46 123.44
*Biscuits
and rolls. 26.62 18.31 20.95 27.08 30.87 38.69 45.14
*Cakes and
cupcakes.. 20.31 13.30 15.37 21.50 26.94 26.92 31.29
*Bread and
cracker
products.. 2.82 2.70 2.40 2.36 3.73 4.07 4.89
*Sweetrolls
, coffee
cakes,
doughnuts. 19.60 15.04 15.68 18.44 23.22 23.31 33.90
*Pies,
tarts,
turnovers. 5.48 5.38 5.07 4.76 8.16 6.48 8.22
*Meats,
poultry, fish
and eggs...... 560.01 477.38 555.07 541.91 635.94 699.55 812.35
*Beef........ 183.66 152.35 204.56 185.96 215.42 225.57 263.75
*Ground
beef...... 79.09 71.32 84.22 79.31 96.47 91.12 101.79
*Roast....... 33.40 28.09 34.54 34.19 35.85 40.13 50.81
*Chuck
roast..... 13.23 11.36 14.43 13.12 16.66 14.71 17.87
*Round
roast..... 9.13 7.79 10.19 8.42 8.80 15.67 13.71
*Other
roast..... 11.04 8.93 9.92 12.65 10.28 9.75 19.23
*Steak....... 59.01 41.47 71.43 61.32 70.25 77.08 93.67
*Round
steak..... 11.62 11.60 16.74 13.83 12.33 11.66 13.85
*Sirloin
steak..... 12.96 8.51 11.79 12.72 14.53 20.48 24.43
*Other
steak..... 34.42 21.36 42.90 34.77 43.40 44.94 55.39
*Other beef.. 12.17 11.47 14.37 11.14 12.84 17.24 17.48
*Pork.......... 114.19 104.51 108.16 108.24 132.60 131.31 157.61
*Bacon....... 20.23 24.20 17.46 18.44 23.15 18.48 25.09
*Pork chops.. 27.10 19.23 28.84 25.14 36.30 28.37 34.33
*Ham......... 27.43 25.79 27.94 28.11 31.54 35.35 39.08
*Ham, not
canned.... 24.47 21.68 25.11 26.25 28.62 29.58 36.43
*Canned ham 2.96 41.0 2.83 1.86 2.92 5.77 2.65
*Sausage..... 16.60 14.09 17.07 13.98 17.67 22.46 23.28
*Other pork.. 22.83 21.21 16.84 22.58 23.95 26.65 35.83
*Other meats... 83.61 71.60 75.58 79.96 98.98 113.62 118.21
*Frankfurters 17.37 15.97 17.17 17.77 19.56 21.76 22.58
*Lunch meats
(cold cuts). 58.88 49.13 49.48 56.28 71.83 80.88 86.21
*Bologna,
liverwurst
, salami.. 19.11 17.97 16.54 19.36 22.06 23.99 25.51
*Other
lunchmeats 39.78 31.16 32.94 36.92 49.77 56.89 60.70
*Lamb, organ
meats and
others...... 7.36 6.51 8.93 5.91 7.58 10.98 9.42
*Lamb and
organ
meats..... 6.17 5.97 5.33 4.82 7.57 8.63 8.50
*Mutton,
goat and
game...... 1.19 0.54 3.60 1.10 0.01 2.36 0.92
*Poultry....... 85.49 69.40 81.53 82.16 85.67 111.40 133.20
*Fresh and
frozen
chickens.... 66.41 55.25 66.61 65.26 66.71 81.88 96.35
*Fresh
whole
chicken... 17.24 17.03 17.44 20.11 14.32 16.24 23.38
*Fresh and
frozen
chicken
parts..... 49.17 38.22 49.18 45.15 52.39 65.64 72.97
*Other
poultry,
incl whole
frzn
chickens.... 19.08 14.15 14.91 16.91 18.96 29.52 36.85
*Fish and
seafood....... 65.24 50.44 57.57 56.89 74.24 86.03 109.89
*Canned fish
and seafood. 17.95 14.01 15.77 16.67 23.21 21.96 26.53
*Fresh and
frozen
shellfish... 14.98 5.89 17.35 14.09 15.81 18.33 32.78
*Fresh and
frozen
finfish..... 32.31 30.53 24.45 26.13 35.23 45.74 50.59
*Eggs.......... 27.83 29.08 27.68 28.69 29.02 31.61 29.68
*Dairy products.. 277.91 237.49 246.39 287.05 337.97 365.06 383.11
*Fresh milk and
cream......... 134.41 132.08 125.44 135.91 160.12 158.15 168.53
*Whole milk.. 52.12 58.46 57.48 57.54 55.58 44.53 55.37
*Other milk
and cream... 82.29 73.61 67.97 78.37 104.54 113.61 113.17
*Other dairy
products...... 143.50 105.41 120.95 151.15 177.85 206.92 214.57
*Butter...... 8.89 8.28 8.17 8.63 9.20 12.22 13.04
*Cheese...... 79.01 54.41 69.16 83.03 98.98 111.72 119.08
*Ice cream
and related
products.... 41.68 31.02 33.95 46.55 53.10 60.05 61.85
*Miscellaneou
s dairy
products.... 13.93 11.70 9.67 12.93 16.57 22.93 20.60
*Fruits and
vegetables...... 376.38 327.70 335.02 366.35 441.76 487.04 526.17
*Fresh fruits.. 120.98 102.64 104.99 116.33 148.47 156.62 172.21
*Apples...... 21.57 16.83 17.80 21.56 26.58 28.77 29.44
*Bananas..... 20.65 19.42 19.04 21.64 24.06 22.66 27.22
*Oranges..... 10.98 9.36 9.43 9.37 15.38 16.47 14.72
*Other fresh
fruits...... 67.78 57.02 58.71 63.75 82.45 88.72 100.83
*Fresh
vegetables.... 110.67 101.90 100.87 106.30 124.19 123.40 158.76
*Potatoes.... 16.61 13.93 17.56 15.59 19.18 20.55 22.24
*Lettuce..... 13.73 11.36 11.61 12.80 16.85 16.95 20.85
*Tomatoes.... 14.87 13.49 14.08 14.64 17.94 16.43 19.74
*Other fresh
vegetables.. 65.47 63.12 57.63 63.28 70.21 69.47 95.93
*Processed
fruits........ 86.81 75.04 80.64 82.22 98.04 126.19 121.27
*Frozen
fruits and
fruit juices 19.59 17.95 18.61 18.42 22.98 32.00 28.71
*Frozen
orange
juice..... 14.43 13.60 13.91 14.55 14.89 21.97 20.88
*Other
frozen
fruits and
juices.... 5.16 4.35 4.70 3.87 8.10 10.03 7.83
*Canned and
dried fruits 21.22 18.05 18.46 20.85 24.80 32.50 26.30
*Fresh,
canned or
bottled
fruit juices 46.00 39.04 43.58 42.95 50.26 61.68 66.26
*Processed
vegetables.... 57.92 48.13 48.52 61.49 71.05 80.82 73.92
*Frozen
vegetables.. 21.30 13.63 16.57 23.19 27.40 32.62 31.07
*Canned and
dried
vegetables
and juices.. 36.62 34.50 31.95 38.30 43.65 48.21 42.85
*Canned
beans..... 6.64 5.41 6.00 6.98 7.85 9.98 7.27
*Canned
corn...... 4.21 2.91 3.86 4.70 4.31 6.10 4.37
*Other
canned and
dried
veg., &
juices.... 25.77 26.17 22.09 26.62 31.49 32.12 31.21
*Other food at
home............ 645.61 500.46 543.39 658.15 765.69 937.83 938.05
*Sugar and
other sweets.. 80.66 65.44 64.53 82.49 97.73 122.23 111.61
*Candy and
chewing gum. 45.41 32.09 31.13 46.37 55.74 75.32 71.53
*Sugar....... 17.07 18.57 17.60 17.61 18.27 17.57 16.01
*Artificial
sweeteners.. 2.36 1.56 2.44 2.90 1.78 3.70 2.68
*Jams,
preserves,
other sweets 15.82 13.22 13.36 15.62 21.94 25.64 21.39
*Fats and oils. 56.65 48.51 45.63 59.62 69.18 76.24 70.54
*Margarine... 11.96 10.65 9.89 12.19 14.51 15.78 15.49
*Other fats,
oils, and
salad
dressing.... 31.66 26.57 26.33 32.32 37.90 45.04 38.58
*Nondairy
cream and
imitation
milk........ 4.49 4.53 3.64 4.91 4.72 5.00 4.89
*Peanut
butter...... 8.54 6.75 5.77 10.20 12.05 10.42 11.57
*Miscellaneous
foods......... 272.98 209.21 230.18 278.73 325.17 410.76 393.38
*Frozen
prepared
foods....... 46.13 34.31 44.35 47.46 54.87 69.97 66.80
*Frozen
meals..... 16.75 14.44 19.43 16.05 23.09 21.11 23.97
*Other
frozen
prepared
foods..... 29.39 19.87 24.92 31.41 31.78 48.86 42.82
*Canned and
packaged
soups....... 21.41 17.65 16.96 21.06 24.10 35.62 28.50
*Potato
chips, nuts,
and other
snacks...... 59.78 41.00 37.67 64.36 71.49 95.82 100.20
*Potato
chips and
other
snacks.... 46.79 30.06 31.54 53.75 55.18 74.06 77.26
*Nuts...... 12.99 10.94 6.13 10.61 16.31 21.76 22.94
*Condiments
and
seasonings.. 61.52 49.20 56.11 58.41 77.90 82.89 92.16
*Salt,
spices,
other
seasonings 12.31 10.17 11.24 11.99 14.15 14.15 20.09
*Olives,
pickles,
relishes.. 7.62 5.36 8.73 7.01 9.48 10.54 10.59
*Sauces and
gravies... 31.62 25.10 26.76 29.73 42.60 43.35 46.87
*Baking
needs and
misc.
products.. 9.97 8.58 9.38 9.67 11.67 14.85 14.61
*Other canned/
packaged
prepared
foods....... 84.14 67.05 75.08 87.44 96.82 126.46 105.73
*Salads and
desserts.. 13.23 11.82 12.59 11.96 17.20 18.07 18.37
*Baby food. 16.25 10.38 15.03 16.91 19.52 26.99 12.93
*Miscellane
ous
prepared
foods..... 54.66 44.85 47.47 58.57 60.10 81.41 74.43
*Nonalcoholic
beverages..... 204.37 164.51 186.49 210.29 233.06 283.11 287.11
*Cola........ 92.19 66.57 88.15 99.91 101.86 140.51 123.90
*Other
carbonated
drinks...... 32.62 23.55 29.15 28.70 40.86 43.90 53.99
*Coffee...... 40.93 38.84 38.48 38.15 43.16 47.73 54.95
*Roasted
coffee.... 25.27 22.96 23.27 24.63 26.36 31.15 34.98
*Instant
and freeze
dried
coffee.... 15.66 15.87 15.21 13.52 16.81 16.58 19.98
*Non-
carbonated
fruit
flavored
drinks...... 16.30 12.30 11.92 21.28 20.62 24.65 21.32
*Tea......... 11.18 10.67 8.25 11.36 13.49 15.33 13.42
*Other non-
alcoholic
beverages... 11.15 12.58 10.54 10.90 13.08 10.98 19.52
Food prepared
by cu on out
of town trips. 30.94 12.80 16.56 27.01 40.55 45.49 75.42
Food away from home 1627.45 968.10 1240.03 1591.02 2030.75 2375.06 3186.24
*Meals at
restaurants,
carry-outs &
other........... 1275.77 799.32 1039.21 1294.24 1591.66 1870.30 2351.22
*Lunch......... 499.88 277.04 407.25 514.76 619.15 709.45 956.78
*Dinner........ 549.30 339.39 440.28 550.06 662.77 822.65 1057.00
*Snacks and non
alcoholic
beverage...... 142.56 105.40 121.94 145.83 190.00 225.33 207.78
*Breakfast and
brunch........ 84.04 77.48 69.75 83.59 119.74 112.87 129.66
Board (including
at school)...... 43.62 6.74 7.89 27.65 36.46 39.33 153.00
Catered affairs 41.27 7.39 5.78 34.97 50.79 47.01 142.76
Food on out of
town trips.... 195.31 93.30 115.14 165.61 254.20 300.02 451.05
School lunches. 42.24 20.43 26.30 41.51 67.39 84.77 70.55
Meals as pay... 29.24 40.92 45.71 27.04 30.27 33.64 17.65
Alcoholic beverages 281.70 182.87 235.22 290.56 343.77 352.96 506.47
*At home......... 148.36 107.27 126.68 152.37 189.69 178.29 246.36
*Beer and ale.. 89.05 72.34 77.77 95.86 108.21 102.60 126.68
*Whiskey....... 12.73 12.89 5.93 13.17 16.76 13.43 21.68
*Wine.......... 32.15 13.69 26.16 31.70 40.94 46.88 70.20
*Other
alcoholic
beverages..... 14.43 8.35 16.82 11.65 23.78 15.38 27.80
Away from home... 133.34 75.61 108.54 138.19 154.08 174.67 260.11
*Beer and ale.. 37.50 20.21 32.77 39.59 40.20 53.06 62.61
*Wine.......... 18.54 12.05 15.72 19.17 18.68 24.52 38.32
*Other
alcoholic
beverages..... 58.12 36.45 50.81 60.89 71.66 67.59 113.53
Alcoholic
beverages
purchased on
trips......... 19.17 6.90 9.24 18.54 23.54 29.50 45.66
Housing............ 8069.13 5495.09 5946.80 7511.85 9260.40 10608.79 15719.12
Shelter.......... 4470.25 3043.10 3139.50 4124.86 5049.86 5901.40 8909.44
Owned dwellings 2554.04 961.15 1151.03 1976.74 2970.57 4060.42 6925.93
Mortgage
interest.... 1560.48 318.45 520.13 1051.78 1925.39 2783.87 4724.67
Mortgage
interest
and
charges... 1560.38 318.45 520.13 1051.78 1925.39 2783.87 4724.01
Prepayment
penalty
charges.
(own home) 0.10 0.00 0.00 0.00 0.00 0.00 0.66
Property
taxes....... 496.08 316.48 301.71 417.03 599.30 643.81 1125.91
Maintenance,
repairs,
insur, othr
expenses.... 497.48 326.23 329.20 507.92 445.87 632.74 1075.35
Homeowners
and
related
insurance. 151.74 102.72 105.11 139.48 163.01 200.84 313.16
Fire and
extended
coverage 4.98 3.14 4.17 8.42 3.90 7.28 4.94
Homeowner
s
insuranc
e....... 146.76 99.58 100.94 131.06 159.11 193.56 308.22
Ground rent.. 26.88 26.40 38.12 35.13 23.98 14.38 12.57
Maintenance
and repair
service..... 252.68 166.84 159.60 260.79 187.82 293.13 607.16
Painting
and
papering.. 52.01 34.57 12.75 55.29 21.09 49.82 144.99
Plumbing
and water
heating... 23.06 12.17 16.84 22.72 17.23 32.94 55.39
Heat, a/c,
electrical
work...... 42.03 40.31 20.65 28.44 44.68 61.59 90.02
Roofing and
gutters... 46.96 21.78 51.88 49.75 40.70 55.18 111.23
Other
repair/mai
ntenance
service... 78.78 53.15 53.69 101.29 54.90 85.14 164.91
Repair &
replace
hard
surface
flooring.. 8.14 4.68 2.92 2.46 7.98 6.77 33.94
Repair of
built-in
appliances 1.68 0.16 0.88 0.85 1.24 1.69 6.68
Maintenance/r
epair commod 65.41 28.29 25.78 71.76 70.93 122.37 141.50
Paints,
wallpaper
and
supplies.. 17.47 6.93 5.76 14.64 18.25 33.17 45.07
Tools and
equipment
for
painting
and
wallpaperi
ng........ 1.88 0.74 0.62 1.57 1.96 3.56 4.84
Plumbing
supplies
and
equipment. 5.65 2.25 3.48 6.92 6.24 11.04 11.40
Electrical
supplies,
heat/cool
equip..... 3.76 0.62 4.14 3.32 4.94 2.21 10.24
Materials
for hard
surface
floor,
repair and
replace... 1.85 0.82 0.03 1.32 0.66 5.63 5.85
Material
and
equipment
for roof/
gutters... 5.18 3.60 3.34 8.46 4.25 3.31 5.16
Materials
for
plaster,
panel,
siding,
windows,
doors,
screens,
awnings... 11.08 9.36 4.91 12.62 12.57 15.49 23.43
Materials
for patio,
walk,
fence,
drive,
masonry,
brick, and
stucco
work...... 2.12 0.28 0.61 6.99 0.71 2.30 3.03
Materials
for
landscapin
g
maintenanc
e......... 2.52 0.09 0.00 7.40 2.38 2.17 4.45
Miscellaneo
us
supplies/e
quipment.. 13.89 3.60 2.89 8.52 18.98 43.49 28.02
Materials
for
insulati
on,
other
maintena
nce/repa
ir...... 7.87 3.60 2.36 6.16 10.04 14.88 18.16
Materials
to
finish
basement
,
remodel
rms or
build
patios,
walks,
etc
(maint.,
rep.,
repl.)
(own
prop)... 6.02 0.00 0.53 2.36 8.95 28.61 9.86
Property
management and
security........ 0.74 1.98 0.53 0.72 0.13 1.95 0.79
Property
management.. 0.64 1.88 0.24 0.60 0.12 1.95 0.62
Management
and upkeep
serv for
security.... 0.10 0.10 0.28 0.13 0.01 0.00 0.18
Parking........ 0.04 0.00 0.06 0.03 0.00 0.07 0.17
Rented dwellings. 1469.41 1753.31 1777.24 1804.99 1563.71 1248.94 825.42
Rent........... 1428.30 1708.38 1718.30 1762.19 1521.88 1216.05 785.55
Rent as pay.... 17.34 25.29 32.67 15.87 14.30 0.70 9.59
Maintenance,
insurance and
other expenses 23.76 19.63 26.2

[Text truncated at 120,000 characters. The full text is on the page linked above.]

---

Source: Frix Law Library, https://www.frixlaw.com/law-library/documents/fr%3A94-12463. Public record. Not legal advice.