Cost-of-Living Allowances (Nonforeign Areas); Final Rule

OFFICE OF PERSONNEL MANAGEMENT

5 CFR Part 591

RIN 3206-AF52

Cost-of-Living Allowances (Nonforeign Areas)

AGENCY: Office of Personnel Management.

ACTION: Final rule.

-----------------------------------------------------------------------

SUMMARY: The Office of Personnel Management (OPM) is issuing final

regulations to increase certain cost-of-living allowance (COLA) rates

paid to General Schedule, U.S. Postal Service, and certain other

Federal employees in Guam and the Commonwealth of the Northern Mariana

Islands; the City and County of Honolulu, Hawaii; and St. Thomas and

St. John, Virgin Islands. The increases are based on living cost

surveys conducted by Runzheimer International, under contract with OPM,

during the summer of 1992 and winter of 1993.

EFFECTIVE DATES: These regulations are effective March 23, 1994, and

are applicable on the first day of the first pay period beginning on or

after March 23, 1994.

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

SUPPLEMENTARY INFORMATION: Under section 5941 of title 5, United States

Code, certain Federal employees in nonforeign areas outside the 48

contiguous States are eligible for cost-of-living allowances (COLAs)

when local living costs are substantially higher than those in

Washington, DC. Currently, nonforeign area COLAs are paid in the

following locations: Alaska, Hawaii, Puerto Rico, the U.S. Virgin

Islands, and Guam and the Commonwealth of the Northern Mariana Islands.

OPM contracted with Runzheimer International to conduct living cost

surveys in the allowance areas in 1992 and 1993. All allowance areas,

except those in Alaska, were surveyed in the summer of 1992. Alaska was

surveyed during the winter of 1993.

The surveys showed that adjustments in various COLA rates were

warranted. This included increases of three COLA rates in three

allowance areas and reductions of eight COLA rates in six allowance

areas. However, a provision in the Treasury, Postal Service, and

General Government Appropriations Act of 1992 (Public Law 102-141) bars

OPM from reducing any COLA rate through December 31, 1995. Therefore,

only the COLA rate increases will be implemented.

The increases implemented by this rulemaking are summarized in the

following table:

Increases in COLA Rates

------------------------------------------------------------------------

Current Final

Allowance area/category rate rate

------------------------------------------------------------------------

City and County of Honolulu, Hawaii Commissary/

Exchange........................................... 15.0 17.5

Territory of Guam and Commonwealth of the Northern

Mariana Islands:

Local Retail...................................... 15.0 22.5

Commissary/Exchange............................... 7.5 17.5

St. Thomas and St. John, Virgin Islands all

employees.......................................... 15.0 17.5

------------------------------------------------------------------------

On August 30, 1993, OPM published proposed regulations (58 FR

45556) that would effect the above increases in COLA rates. On the same

day, OPM published a notice (58 FR 45558) that included Runzheimer's

``Report to OPM on Living Costs in Selected Nonforeign Areas and in the

Washington, DC Area, May 1993.'' In response to the proposed

regulations and notice, OPM received comments from nearly 250 persons.

An analysis of the comments follows.

General Comments

One commenter stated that OPM did not comply with provisions of the

Treasury, Postal, and General Government Appropriations Act 1992 (Pub.

L. 102-141) as these provisions apply to COLA. This law requires that

OPM study living cost issues and submit to Congress a report on

possible changes in the COLA methodology. The report is due March 1,

1995. The commenter believes that the law directs OPM to make changes

in the COLA model before 1995.

OPM's General Counsel carefully reviewed Pub. L. 102-141 and the

related Senate Appropriations Committee report. The General Counsel

determined that the law has two requirements: (1) COLA rates may not be

reduced through December 31, 1995, and (2) OPM must submit a report to

Congress on possible changes to the COLA methodology. The law does not

direct OPM to implement methodological changes at this time.

The Senate Committee, however, requested that OPM research specific

methodological issues. This OPM is doing, and OPM plans to include the

results of this research in its report to Congress. Although the law

does not require OPM to implement changes, OPM will continue to make

improvements in the COLA methodology as appropriate.

Another commenter said that OPM regulations should describe in

greater detail the COLA model and survey. The commenter also stated

that all of the data collected should be made public.

OPM believes that its COLA regulations are adequately detailed and

that any attempt to subject the COLA survey process to a set of overly

detailed and inflexible rules would impair rather than improve the COLA

program. The flexibility results in a more accurate COLA model because

improvements can be made from one year to the next. Such changes are

made public because, before COLA rates are adjusted, OPM publishes in

the Federal Register a detailed report on the survey methodology and

results. Therefore, employees have the opportunity to comment on any

changes.

OPM does not publish all of the raw data collected in the survey

because of the tremendous volume of data. As Runzheimer stated in its

report, over 16,000 price quotes were collected. The report, however,

provides detailed information on the results of the survey. In

addition, OPM provides to those who request it additional data to the

extent authorized under the Freedom of Information Act.

One commenter referred to the confidentiality statement on the

information collection materials that was included in Appendix 5 of the

report. The confidentiality statement says that the Government will

hold all micro or ``raw'' data in confidence. The commenter stated that

OPM should not keep secret the survey data that it collects.

Runzheimer inadvertently included an obsolete version of the

information collection materials in Appendix 5. This version was not

used in the surveys. OPM's policy is to release ``raw'' or micro data

to the extent authorized under the Freedom of Information Act. In

conducting OPM's surveys, Runzheimer does not pledge to hold

confidential the survey data unless such data are covered by the

provisions of the Freedom of Information Act that allow certain data to

be held in confidence.

One commenter stated that the model was inappropriate and said that

OPM should measure differences in levels of living rather than

differences in prices.

Comparison of levels of living implies comparing lifestyles and

that involves comparing differences in needs and preferences. This is a

highly subjective area because one person's ``need'' might be another

person's luxury. OPM is examining this complex issue and plans to

include a discussion of it in its report to Congress.

One commenter stated that OPM should include a component to

compensate employees for the allowance area's remoteness and isolation.

He suggested that 5 percentage points be added to all COLA rates to

reflect intangible living costs caused by remoteness and isolation. The

commenter did not provide an example of an intangible living cost.

OPM does not know what the commenter meant by the term ``intangible

living costs.'' The commenter may have been referring to monetary costs

that are difficult to measure, or he may have been referring to

nonmonetary factors, such as hardship and inconvenience.

OPM believes that the COLA model adequately measures differences in

monetary costs, although improvements and refinements in the model may

be possible. For example, OPM is researching certain additional items,

particularly those that might be purchased more frequently in remote

areas. These items include air transportation, out-of-area college and

university education, and extraordinary medical expenses. OPM is

looking at ways the tangible cost of these items might be included in

the COLA model and plans to address this issue in its report to

Congress.

OPM believes, however, that nonmonetary factors, such as hardship

and inconvenience, should not be part of the COLA program. There are

other programs that compensate Federal employees for such

circumstances.

One commenter maintained that an item needed in an allowance area,

but not needed in Washington, DC, should be priced only in the

allowance area. The commenter said that the frequency of need also

should be a factor.

Generally, the model compares the cost of an item in an allowance

area with the cost for the same item in the DC area. OPM believes that

this is consistent with the settlement of Hector Arana, et al. v.

United States, in which the plaintiffs asked that OPM adopt a

methodology that compared specified brands, models, and sizes whenever

possible.

Nevertheless, the COLA model does reflect some differences between

areas. For example, the model assumes that cars in Alaska have certain

accessories, such as engine block heaters, that are not common in the

DC area. Also, differences in home construction (e.g., triple pane

windows and greater wall insulation common in Alaska) are included in

the model to the extent that these differences are reflected in real

estate prices. OPM is researching related issues and plans to address

them in its report to Congress.

A few commenters objected to the use of national consumer

expenditure patterns in the living cost model. The commenters believed

that local consumption patterns should be used. More than one commenter

noted that the spending pattern data were old.

To compare living costs between areas, Runzheimer assigned a common

set of weights to each item, category, and component. These weights

reflect how consumers spend their money and were used to derive

comparative indices measuring overall living costs. Runzheimer used

Bureau of Labor Statistics (BLS) nationwide Consumer Expenditure Survey

(CES) data for these weights.

As discussed in the report, the COLA model uses an indexing

methodology similar to the Laspeyres index. As the report also notes,

it would be preferable to use Washington, DC, consumer expenditure data

with the Laspeyres approach. Washington, DC, CES data, however, are not

available by income level, and OPM regulations require measurement of

living costs at multiple income levels. On the other hand, nationwide

CES data are arrayed by income level. Therefore, Runzheimer used these

data in the COLA model.

CES data are also available for Honolulu and Anchorage; but as with

the Washington, DC, data, the Honolulu and Anchorage data are not

available by income level. BLS CES data are not available for any other

nonforeign area (outside the 48 contiguous States), and OPM knows of no

other source of comprehensive consumer expenditure information by

income level suitable for use in the COLA model. Therefore, the use of

local weights is not practical.

Both OPM and Runzheimer recognize that the CES data are old. OPM is

developing a methodology to introduce gradually more recent CES data

into the model. OPM plans to use this approach beginning with the

surveys to be conducted in the summer of 1994.

One commenter suggested that the COLA model be simplified to use

only one income level. The commenter believed that using only one

income level would reduce survey costs and the number of subjective

assumptions required.

As noted earlier, OPM's regulations require the measurement of

living costs at multiple income levels. This approach recognizes that

relative living costs may vary by income level and that the

distribution of employees by income level may vary among areas.

Therefore, the multiple income approach yields a more accurate measure

of overall living cost differences than a single income approach.

Nevertheless, to the extent that multiple income levels require

additional subjective assumptions, the overall integrity of the model

might not be impaired by using a single income level. OPM is examining

this issue and plans to include its findings in the report to Congress.

One commenter objected to Runzheimer's recommendation that OPM

include income taxes in the COLA model. The commenter believed that

this would unduly complicate the model and argued that it would also be

necessary to compare the level of government services available in each

area. Another commenter, however, stated that income taxes were high in

Hawaii and recommended that income taxes be included in the model.

OPM is studying Runzheimer's recommendation and issues relating to

Federal, State, and local income taxes and plans to include the results

of this study in its report to Congress.

A few commenters seemed to have confused the annual living cost

surveys with the special Federal Employee Housing and Living Patterns

Survey, which OPM conducted in the winter of 1992/1993. The commenters

said they could not see how Runzheimer had incorporated the results of

the employee survey in calculating living cost indices.

As OPM stated in the preface to the employee survey, the purpose of

the survey was to collect information that would be used to improve the

COLA model. The preface made it clear that the results of the survey

would not be used directly to set COLA rates.

OPM is now in the process of analyzing the results of the employee

survey. It is expected that these analyses will allow OPM to identify

the subdivisions and communities in which Federal employees live, the

types of housing expenses they incur, the kinds of stores they

frequent, their transportation needs, and so on. OPM plans to use this

information in the design of future COLA surveys to reflect more

closely Federal employee living costs. The information will also aid

OPM as it studies the COLA methodology and prepares its report to

Congress.

Comments on the Goods and Services Component

A commenter from Hawaii said that Federal employees on Oahu did not

make many catalog purchases. In contrast, an employee from Maui stated

that she made many catalog purchases. Likewise, a St. Croix resident

wrote that Virgin Island employees made frequent catalog purchases.

OPM asked Runzheimer to include a limited number of catalog items

in the survey because catalogs are a common source of retail goods and

are used by many persons in all areas, including the Washington, DC,

area. Of course, all catalog prices surveyed included shipping costs

and any applicable local sales and excise taxes. Catalog pricing also

allows better comparisons of items that would otherwise be difficult to

compare. For example, some furniture items were priced in catalogs

because finding comparable styles, brands, and models in earlier

surveys proved difficult.

In the employee survey, OPM asked Federal employees about their

purchasing patterns including whether they typically purchased various

types of items by catalog. OPM plans to use this information in

designing future surveys and in its report to Congress on possible

changes in the COLA methodology.

One commenter believed that catalog pricing understated price

differences between the allowance area and the Washington, DC, area.

The commenter said that in the DC area consumers could buy an item

locally if catalog prices were relatively high but that in the

allowance areas consumers frequently did not have that choice.

Many of the items that Runzheimer priced by catalog are not sold

locally unless the catalog retailer also has a local retail outlet and

that outlet carries the same item. If the item is sold locally by the

retailer, it is usually sold at a price comparable to the catalog

price, unless the item is on sale in either the catalog or in the

retail store. Since Runzheimer does not survey sale prices, the use of

catalog pricing probably does not cause bias.

Another commenter questioned whether representative types of stores

were surveyed in the allowance areas. He believed that stores

frequented in the allowance areas could be significantly different from

those frequented in the Washington, DC, area. He also stated that there

was only one ``warehouse-type'' grocery store on Oahu and that, because

this outlet was less accessible than others, it was inappropriate to

include it in the survey.

Runzheimer surveyed prices at the largest, most popular stores in

each area. These stores included major grocery stores, department

stores, discount stores, and specialty stores. OPM believes that this

process is objective and leads to an equitable comparison of typical

prices between areas.

In addition, Runzheimer selects outlets that are apt to be

frequented by residents of the living communities in which housing is

surveyed. At times, a balance between the types of outlets and their

proximity to certain living communities is difficult to achieve. During

the 1992 survey on Oahu, the grocery store in question was included in

the survey. Subsequently, it was determined that this store was

probably located outside the area normally frequented by residents of

the living communities covered by the survey. Therefore, for the 1993

survey a different outlet was selected to replace the one in question.

Since there were no other ``warehouse-type'' outlets on Oahu, the new

outlet was a conventional, large supermarket.

In the employee survey, OPM asked Federal employees where they

lived, where they shopped, what they purchased, and so on. The survey

included specific questions on the kinds of stores employees

frequented. OPM plans to use the results of the employee survey to

review outlet selection and make changes as appropriate.

One commenter criticized Runzheimer for not considering the cost of

college and university education in the living cost surveys. The

commenter stated that due to limited post-high school educational

opportunities in the allowance areas, Federal employees must send their

children to out-of-area schools.

Runzheimer noted in the report that post-high school educational

opportunities vary significantly among the allowance areas. Most of the

allowance areas, however, have colleges or universities in the major

population areas, and many of these institutions offer a wide range of

degree programs. Nevertheless, Federal employees may send their

children to out-of-area schools.

Without additional information about the frequency of use of

within-area and out-of-area schools, it is not appropriate to include

post-high school education expense in the COLA model. In the employee

survey, OPM asked employees several questions about college and

university usage. OPM plans to use the results of the employee survey

to review the issue of post-high school education. OPM is also

researching the cost of within-area and out-of-area tuition, books,

room and board, transportation, and related expenses. OPM plans to

include the results of this research and the employee survey in its

report to Congress.

Several commenters questioned whether Runzheimer's survey

adequately covered childcare expenses. An employee from Alaska stated

that her childcare costs were high and accounted for a large percentage

of her total budget.

Two kinds of childcare are included in the survey--day care and

babysitting. Runzheimer prices the monthly cost of professional day

care services (eight hours a day, five days a week). Runzheimer also

obtains the price of casual babysitting services. Both are assigned

appropriate weights based on the CES and are used in the COLA model.

Comments on the Housing Component

Some commenters objected to trimming the high and low values in the

housing component. The commenters believed that housing market price

anomalies should be tolerated or that another approach should be used

to reduce these anomalies.

As was stated in the report, the purpose of trimming was to

stabilize the housing data from one year to the next. Trimming is

essentially a nonparametric technique, similar to using the median

rather than the average. OPM and Runzheimer considered using the median

but rejected it because the limited number of observations obtained in

some smaller allowance areas could cause the median to be erratic from

one year to the next. Runzheimer recommended trimming as an alternative

to the median, and OPM agreed. Trimming provides stability; and because

equal numbers of high and low values are trimmed, no bias is

introduced.

Another commenter objected to the comparison of new and older home

prices combined. He felt that the survey should compare the prices of

homes of a similar age as well as a similar size and room count.

Numerous factors influence rents and selling prices. Information on

some of these factors is readily available, but much of it is not.

Runzheimer uses home size and room count as the major criteria in

housing comparisons because these factors generally have the most

influence on housing costs. Age is not used because it frequently is

not available and probably has less influence.

One commenter from Alaska noted the high cost of her new home in

Alaska. She also said that the cost of drilling a deep well

significantly increased the cost of her new home.

Runzheimer surveys the selling price of homes that sold during the

6 month period prior to the survey. The selling price generally

reflects the cost of construction, including the cost of appurtenances

such as water wells.

One commenter stated that the residential areas surveyed on Oahu

did not properly reflect where Federal employees live nor the income

levels that Runzheimer associated with the communities.

OPM recognizes that community selection is an important part of the

COLA survey. The communities surveyed in Hawaii were changed in

response to comments OPM received on earlier surveys. Additional

changes may be warranted. OPM plans to use the results of the employee

survey to review community selection and make appropriate changes.

A commenter from Alaska stated that the cost of utilities was high

and provided examples of her utility costs. She also stated that

utility costs vary with the size of the home.

Runzheimer included in the COLA surveys the cost of utilities. The

average costs for Owners and Renters for each area were shown in

Appendix 7 of the report and were part of the Federal Register notice.

As shown in the appendix, the cost of utilities is the second highest

cost of housing, exceeded only by the cost of mortgage payments or

rent.

The COLA model takes into account that utility costs vary with home

size. Section 4.2.4.1 of the report described the process used and the

factors that were applied.

One commenter stated that the survey failed to take into

consideration the use of solar water heaters in Hawaii and Guam. The

commenter believed that the model did not account for the capital cost

of such heaters nor the possible reduction in overall utility

consumption.

As noted above, significant home features and improvements

generally are reflected in the selling price of the home.

Therefore, OPM's living cost surveys will reflect the cost of solar

water heaters to the extent that such items influence home market

values and are commonly found on homes in any area, including Hawaii

and Guam. If the use of solar water heaters is so common that it

generally reduces the consumption of utilities, this too will be

reflected in the survey results.

This is as it should be. The COLA model compares overall living

costs in the allowance area with overall living costs in the DC area.

If housing is more expensive because solar heaters are common in an

allowance area but not the DC area and if overall utility costs are

lower in the allowance area because solar heaters are used extensively

but are not used in the DC area, the final comparison of overall

housing costs will be equitable. No special consideration of capital

improvement costs or reduced utility consumption is appropriate.

Several commenters noted that employees in the allowance areas face

extreme weather disturbances, particularly typhoons or hurricanes. The

commenters stated that these weather disturbances create higher costs

in home maintenance and insurance.

Runzheimer surveys the cost of home insurance. If insurance costs

increase after a major natural disaster, the COLA surveys will reflect

these higher costs. Other issues, such as the cost of repairing storm

damage, are more difficult to address in the surveys. Although it may

be possible to price the cost of repairing or replacing an item such as

a window or a roof, it is difficult to know how often this must be done

in each allowance area compared with the Washington, DC, area. In the

employee survey, OPM asked about storm damage, home maintenance, and

frequency of repairs. OPM plans to review this issue carefully in light

of the results of the employee survey.

One commenter asserted that Federal employees frequently purchase

disaster insurance (e.g., home insurance covering damage caused by

floods, storms, or earthquakes) and criticized Runzheimer for not

including the cost of these additional insurance riders.

Runzheimer interviews local insurance agents to obtain the cost of

insurance in each area. In these interviews, Runzheimer asked agents

about disaster insurance and whether it was typically purchased by

homeowners in the allowance area. Runzheimer concluded from these

interviews that such insurance is not typically purchased and,

therefore, recommended against including it in the COLA model. OPM

agreed. However, questions regarding disaster insurance were included

in the employee survey. OPM plans to reevaluate this issue in light of

the results of the employee survey and address this issue in its report

to Congress.

The same commenter noted that Runzheimer was unable to obtain the

price of home insurance on Guam because insurance companies had issued

a temporary ``moratorium'' on the sales of new policies after Typhoon

Omar. The commenter criticized Runzheimer for using 1991 survey data in

place of the missing data.

Runzheimer discussed the Guam insurance issue with OPM as soon as

the issue arose. Because there was no indication of the amount of any

forthcoming rate increase or that rates would increase at all,

Runzheimer and OPM believed that it was inappropriate to adjust

artificially the 1991 insurance rates. To the extent insurance

companies adjusted their rates after the moratorium, such rate changes

were obtained in the following Guam survey and will be appropriately

reflected in the results of that survey.

One commenter questioned how Runzheimer obtained survey data in

Kauai after Hurricane Iniki because the Runzheimer researcher from the

central office had not been allowed to visit the island.

Most of the living cost data are obtained by either: (1) Local data

collectors who are residents of the area or (2) by telephone research

conducted from Runzheimer's central office. Senior personnel from

Runzheimer's central office visit the allowance areas to monitor and

review the survey process. These visits are conducted after the on-site

data collection is complete.

In the case of the Kauai survey, all on-site data collection had

been completed prior to the arrival of Iniki. The fact that Runzheimer

officials from the central office were unable to visit the island is

not significant because the data collector had done an excellent job,

and the quality of the data collected was quite good. Because the

prices surveyed were pre-Iniki, they were not influenced by any short-

term perturbations caused by the hurricane. Runzheimer officials were

able to visit Kauai as part of the summer 1993 survey, and a great deal

of attention was given to collecting and analyzing data from that later

survey.

The commenter also stated that because some utilities were not

widely available for an extended period after Hurricane Iniki, the

living cost surveys might show that utility usage was low. He said that

this could bias the survey results.

The utility usage factors that Runzheimer obtained on Kauai were

based on a period prior to Iniki. Therefore, the hurricane did not

distort the survey data. In the conduct of the most recent survey,

Runzheimer paid close attention to utility usage rates to ensure that

the survey results were not unduly influenced by the effects of Iniki.

Several commenters said that climate conditions (such as high

humidity, high rainfall, sunlight intensity, airborne salt, snow, and

cold weather) resulted in higher home maintenance costs in the

allowance areas than in the Washington, DC, area. One commenter

believed that some home maintenance expenses were incurred more

frequently in the allowance areas but that Runzheimer considered only

annual maintenance costs.

In the employee survey, OPM asked several questions concerning home

maintenance, such as painting and roof replacement. OPM is also

studying these issues in a closely related, special research project.

OPM plans to integrate the results of the employee survey with the

special research and include this in its report to Congress.

One commenter from Hawaii stated that leasehold to fee-simple

ownership conversions contributed to higher housing costs in Hawaii.

Runzheimer surveys only fee-simple home sales in Hawaii. Leasehold

properties are excluded. OPM believes that the fair market value of

fee-simple property adequately reflects the market as a whole--both the

leasehold market (in which the homeowner may have to purchase the land

or renegotiate a land lease) and the fee-simple market.

Another commenter expressed concern whether the survey of fee-

simple home sales only resulted in the survey of typical housing. The

commenter suggested that Runzheimer also survey leasehold properties

and include the annual cost of the land lease.

When OPM published previous Runzheimer reports in the Federal

Register, numerous commenters expressed the view that leasehold sales

in Hawaii should not be included in OPM's living cost surveys. For the

reasons discussed above, OPM agrees with this position and has directed

Runzheimer to continue its practice of surveying only fee-simple sales.

Comments on the Transportation Component

A number of commenters stated that private transportation costs

were greater in the allowance areas because of the high cost of

automobiles and increased auto maintenance due to poor roads, rough

terrain, salt air, and harsh weather. Many also felt that their

automobile insurance premiums were quite high. One commenter suggested

that OPM price the cost of tinting car windows.

The COLA model takes into consideration automobile purchase price,

maintenance, insurance, and depreciation. Purchase costs and insurance

are based on price data obtained in each area. Maintenance is also

based on local price data, and the model assumes that certain types of

maintenance occur more frequently in the allowance areas than in the DC

area. For example, the model assumes that tires wear out faster in the

allowance areas than in the Washington, DC, area, and that tires have

to be purchased more frequently in the allowance areas.

Depreciation is based on used car values, and Runzheimer found that

used cars generally depreciate at the same rate in nearly all areas.

The exceptions are Nome and Fairbanks where cars depreciate at a faster

rate, perhaps because of the severe climate. Runzheimer used special

factors for these two areas to reflect greater depreciation.

In the employee survey, OPM asked employees about their car

purchases, accessories, maintenance, road conditions, terrain, and

several other issues. OPM plans to review transportation costs in light

of the results of the employee survey. OPM plans to address items, such

as window tinting, at that time.

Some commenters were confused about the composition of the Public

Transportation Category. Some commenters from the Virgin Islands stated

that the lack of an effective mass transportation system compelled them

to purchase cars or to use taxis.

As explained in the report, Runzheimer surveys airline fares to

determine the cost of Public Transportation. Runzheimer does not survey

municipal mass transportation. The cost of bus, subway, or taxi service

is not part of the surveys because the service available in many

allowance areas is not comparable to the service available in the DC

area. Instead of public mass transportation, Runzheimer compares the

cost of round-trip airfare from the allowance area to Los Angeles,

California, with the cost of round-trip airfare from Washington, DC, to

Los Angeles.

Two commenters objected to the selection of Los Angeles as the

common destination point for comparing airfares. They stated that the

Los Angeles routes were highly competitive and resulted in lower fares

compared with other destinations.

As stated in the report, Los Angeles was selected because it is a

common point within the continental United States that is roughly

equidistant from each of the allowance areas and the Washington, DC,

area. The route may be highly competitive, but that does not invalidate

cost comparisons because it is the relative cost of air travel that is

being measured. If competition reduces fares, the reductions will be

reflected in the Washington, DC, to Los Angeles fares as well as the

allowance area to Los Angeles fares. Therefore, OPM believes the

comparisons are appropriate.

Some commenters stated that the model did not measure true air

transportation costs. The commenters stated that inter-island travel,

travel within Alaska, and travel to the contiguous 48 States required

more frequent use of air transportation.

OPM included in the employee survey several questions regarding

travel. OPM plans to review the transportation component of the COLA

model in light of the results of the employee survey.

Comments on the Miscellaneous Component

One commenter objected to the assumption in the model that the cost

of certain Miscellaneous Component items is the same in the allowance

area as in the Washington, DC, area. The commenter said that cultural

differences might lead to larger expenditures for gifts. The commenter

also noted that the Senate Committee asked OPM to review the

Miscellaneous Component to ensure that the results reflect actual

living costs and do not assume equal costs between areas.

The relative costs of the majority of the items in the

Miscellaneous Component are based on surveyed prices. Therefore, the

Miscellaneous Component index reflects ``actual'' living cost

differences. The cost of only two items does not differ among areas--

(1) Life insurance and pensions and (2) cash contributions and gifts.

For Federal employees, the cost of life insurance and required

contributions to a Federal retirement system do not vary by area. Any

additional insurance or contributions to retirement systems are a

matter of personal preference. Gifts and cash contributions for church,

charity, or other purposes are a matter of personal preference and/or

reflect lifestyle differences that are beyond the scope of the COLA

program. As noted earlier, OPM is studying the issue of lifestyle

differences and plans to discuss it in its report to Congress.

One commenter proposed using the Goods and Services Component index

to adjust the cash contributions/gifts category to reflect the cost of

gift items purchased locally.

OPM is researching this issue along with the general composition of

and assumptions used in the Miscellaneous Component. OPM plans to

include the results of this review in its report to Congress.

One commenter said that the medical expense portion of the

Miscellaneous Component failed to reflect the higher out-of-pocket

expenses that Federal employees in the allowance areas frequently

incurred. The commenter cited several possible causes for such higher

costs including higher costs not covered by insurance carriers, the

absence of Health Maintenance Organizations (HMOs) in several allowance

areas, and the need to travel outside the area to obtain some medical

services.

In the employee survey, OPM asked several questions regarding

medical expenses, and in addition, OPM is researching related health

cost issues. OPM plans to include the results of its research and the

employee survey in its report to Congress.

One commenter stated that employees in the allowance areas have to

save at a higher rate to afford the down payment for a house or car or

to pay for college/university education. The commenter said that OPM

should take this into consideration and use the Goods and Services

Component index to adjust the amount of money saved relative to

Washington, DC.

As noted in the report, Runzheimer believes that savings and

investments made for the purpose of future purchases of housing,

durable goods, education, and similar items are best accounted for in

the category or component associated with the item. OPM agrees with

this approach and notes that this approach is consistent with the

methodology the Bureau of Labor Statistics uses in the CES.

The commenter also stated that the COLA model should take into

consideration the fact that COLAs do not count toward retirement. The

commenter believed that Federal employees had to invest at a higher

rate in pensions and other savings vehicles to afford to retire in the

allowance areas.

Under sections 8331(3) and 8401(4) of title 5, United States Code,

allowances (which includes COLAs) are excluded from basic pay in the

computation of Federal annuities under the Civil Service Retirement

System and the Federal Employees' Retirement System. It would be

inappropriate to adjust COLA to take into consideration that which the

law has specifically excluded. Therefore, OPM believes that no

adjustments to the pensions and investments portion of the model are in

order.

Comments About the Virgin Islands Surveys

A number of employees from the Virgin Islands felt that the COLA

surveys did not accurately reflect living costs, particularly in St.

Croix. The employees said that the COLA rates were too low. One

commenter questioned the validity of the price data collected in St.

Croix.

OPM closely monitors Runzheimer's work and believes that the

surveys and analyses are accurate. OPM specifically reviewed in great

detail all survey data from the Virgin Islands. We are satisfied that

Runzheimer followed appropriate procedures in collecting data,

analyzing, and reporting data.

Some commenters from St. Croix referred to a Virgin Island

Department of Labor survey that indicated that food costs in St. Croix

were 25 percent higher than food costs in the Washington, DC, area. The

employees cited this as evidence that the St. Croix COLA rate should be

higher.

Runzheimer priced a wide variety of food items in each allowance

area, including St. Croix. The results of the food portion of the

survey were provided in Appendix 4 of the report and were part of the

Federal Register notice. These results showed that food consumed at

home was approximately 28 percent more expensive in St. Croix than in

the DC area.

COLA rates, however, are based on more than just the relative cost

of food; and in St. Croix, the relative costs of other items were

generally lower than the relative cost of food. Therefore, the St.

Croix COLA rate is lower than the food index.

Over 200 employees from St. Croix, stated that their COLA rate

should be the same as the rate for St. Thomas and St. John.

The COLA surveys for the two areas showed that some prices were

higher in St. Croix than in St. Thomas and that some were lower.

Overall, St. Croix prices were about 4 percentage points lower than St.

Thomas prices. The difference in the final COLA rates for the two areas

generally reflects this overall price difference.

OPM notes that the American Chamber of Commerce Research

Association (ACCRA) surveyed living costs in the Virgin Islands in

1992. The results of the ACCRA survey also showed that living costs in

St. Thomas were higher than living costs in St. Croix.

One commenter stated that the Virgin Island COLA surveys should not

be conducted during the summer. He maintained that pricing in the

summer reflected lower, off-season prices.

OPM recognizes that survey timing is an important consideration,

and COLA surveys are scheduled to collect prices in a ``typical''

month. OPM believes that the current surveys are conducted at a

reasonable time of year but will consider timing changes. Survey timing

will be part of OPM's report to Congress.

Several commenters from the Virgin Islands stated that certain

medical services were not available in their area and that they had to

fly to other areas to obtain these services. One commenter from Alaska

also noted this problem. In addition, many commenters in St. Croix

stated that the local hospital was not accredited. They said they had

to fly to Puerto Rico or to the U.S. mainland for hospital services.

OPM is studying the availability and cost of medical services in

the allowance areas. In addition, OPM's employee survey included

questions regarding where Federal employees obtain medical services.

OPM plans to include the results of its research and the employee

survey in its report to Congress.

Many employees in St. Croix cited the high cost of air travel for

medical treatment. They also noted the cost of air ambulance service.

As part of its research, OPM is studying the cost of obtaining

medical services in the appropriate area if such services are not

available locally. OPM is also researching the issue of air ambulance

insurance. OPM plans to include the results of this research in its

report to Congress.

Many commenters from St. Croix criticized the quality of public

schools in their area and said that a high percentage of Federal

employees sent their children to private schools. The commenters

believed that OPM should consider the cost of private education in the

survey.

OPM is studying private education issues. In addition, OPM asked

employees in the employee survey whether they sent their children to

public or private schools. OPM plans to include the results of this

research in its report to Congress.

Some employees in St. Croix want OPM to take into account the cost

of sending children to out-of-area colleges and universities. They

noted the high cost of travel, campus housing, and out-of-state

tuition.

OPM is studying the cost of college and university education, and

the employee survey included questions concerning college and

university education. OPM plans to include the results of its research

and the employee survey in its report to Congress.

Regulatory Flexibility Act

I certify that this regulation will not have a significant economic

impact on a substantial number of small entities because the regulation

will affect only Federal agencies and employees.

List of Subjects in 5 CFR Part 591

Government employees, Travel and transportation expenses, Wages.

U.S. Office of Personnel Management.

James B. King,

Director.

Accordingly, OPM is amending 5 CFR part 591 as follows:

PART 591--ALLOWANCES AND DIFFERENTIALS

Subpart B--Cost-of-Living Allowance and Post Differential--

Nonforeign Areas

1. The authority citation for subpart B of part 591 continues to

read as follows:

Authority: 5 U.S.C. 5941; E.O. 10000, 3 CFR, 1943-1948 Comp., p.

792; E.O. 12510, 3 CFR, 1985 Comp., p. 338.

2. Appendix A of subpart B is revised to read as follows:

Appendix A of Subpart B--Places and Rates at Which Allowances Shall

Be Paid

This appendix lists the places where a cost-of-living allowance has

been approved and shows the allowance rate to be paid to employees

along with any special eligibility requirements for the allowance

payment. The allowance percentage rate shown is paid as a percentage of

an employee's rate of basic pay.

------------------------------------------------------------------------

Authorized

allowance

Geographic coverage/allowance category rate

(percent)

------------------------------------------------------------------------

State of Alaska

City of Anchorage and 50 mile radius by road:

Local retail.............................................. 25.0

Commissary/exchange....................................... 17.5

City of Fairbanks and 50 mile radius by road:

Local retail.............................................. 25.0

Commissary/exchange....................................... 20.0

City of Juneau and 50 mile radius by road:

All employees............................................. 25.0

Rest of the State:

All employees............................................. 25.0

State of Hawaii

City and County of Honolulu:

Local retail.............................................. 22.5

Commissary/exchange....................................... 17.5

County of Hawaii:

All employees............................................. 15.0

County of Kauai:

All employees............................................. 17.5

County of Maui and County of Kalawao:

All employees............................................. 22.5

Territory of Guam and Commonwealth of the Northern Mariana

Islands

All locations:

Local retail.............................................. 22.5

Commissary/exchange....................................... 17.5

Commonwealth of Puerto Rico

All locations:

Local retail.............................................. 10.0

Commissary/exchange....................................... 0.0

The Virgin Islands

St. Croix:

All employees............................................. 12.5

St. Thomas and St. John:

All employees............................................. 17.5

------------------------------------------------------------------------

Definitions of Allowance Categories

The following definitions of the allowance categories identified in

the tables in this appendix shall be used to determine employee

eligibility for the appropriate allowance rate:

------------------------------------------------------------------------

Allowance category Definition

------------------------------------------------------------------------

Local retail....................... This category includes those

employees who purchase goods and

services from private retail

establishments.

Commissary/exchange................ This category includes those

employees who shop at private

retail establishments, but who, as

a result of their Federal civilian

employment, also have unlimited

access to commissary and exchange

facilities. This category is

established only in those

allowance areas that have these

facilities.

------------------------------------------------------------------------

Note: Eligibility for access to military commissary and exchange

facilities is determined by the appropriate military department. If

an employee is furnished with these privileges for reasons

associated with his or her Federal civilian employment, he or she

will have an identification card that authorizes access to such

facilities. Possession of such an identification card--i.e., one

issued by reason of his or her Federal civilian employment--is

sufficient evidence that the employee uses the facilities.

[FR Doc. 94-6810 Filed 3-22-94; 8:45 am]

BILLING CODE 6325-01-P