[Federal Register Volume 71, Number 148 (Wednesday, August 2, 2006)]
[Rules and Regulations]
[Pages 43897-43904]
From the Federal Register Online via the Government Publishing Office [www.gpo.gov]
[FR Doc No: 06-6624]



  Federal Register / Vol. 71, No. 148 / Wednesday, August 2, 2006 / 
Rules and Regulations  

[[Page 43897]]


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

OFFICE OF PERSONNEL MANAGEMENT

5 CFR Part 591

RIN 3206-AK67


Cost-of-Living Allowances (Nonforeign Areas); COLA Rate Changes

AGENCY: Office of Personnel Management.

ACTION: Final rule.

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

SUMMARY: The Office of Personnel Management is issuing final 
regulations to change the cost-of-living allowance rates received by 
certain white-collar Federal and U.S. Postal Service employees in 
Alaska, Hawaii, Guam and the Northern Mariana Islands, Puerto Rico, and 
the U.S. Virgin Islands. The changes are the result of living-cost 
surveys conducted in 2002, 2003, and 2004.

DATES: Effective date: September 1, 2006. Implementation date: First 
day of the first pay period beginning on or after September 1, 2006.

FOR FURTHER INFORMATION CONTACT: Donald L. Paquin, (202) 606-2838; fax: 
(202) 606-4264; or e-mail: [email protected].

SUPPLEMENTARY INFORMATION: Section 5941 of title 5, United States Code, 
authorizes Federal agencies to pay cost-of-living allowances (COLAs) to 
white-collar Federal and U.S. Postal Service employees stationed in 
Alaska, Hawaii, Guam and the Northern Mariana Islands, Puerto Rico, and 
the U.S. Virgin Islands. Executive Order 10000, as amended, delegates 
to the Office of Personnel Management (OPM) the authority to administer 
nonforeign area COLAs and prescribes certain operational features of 
the program. OPM conducts living-cost surveys in each allowance area 
and in the Washington, DC, area to determine whether, and to what 
degree, COLA area living costs are higher than those in the DC area. 
OPM sets the COLA rate for each area based on the results of these 
surveys.

Background

    The 2002 Caribbean surveys were the first OPM conducted using the 
new methodology we adopted pursuant to the stipulation of settlement in 
Caraballo et al. v. United States, No. 1997-0027 (D.V.I), August 17, 
2000. Caraballo was a class-action lawsuit in which the plaintiffs 
contested the methodology OPM used to determine COLA rates. In the 
Caraballo settlement, the parties agreed that if the Government adopted 
and maintained certain changes in the COLA program, the plaintiffs 
would be barred from bringing suit over these issues. The complete 
stipulation for settlement is on OPM's Web site at http://www.opm.gov/oca/cola/settlement.asp.
    Before the settlement, the parties entered into a memorandum of 
understanding under which they engaged in a cooperative process to 
study living-cost and compensation issues. The research was exhaustive 
and covered essentially all aspects of the COLA program. A summary of 
that research is available on OPM's Web site at http://www.opm.gov/oca/cola/research.asp.
    Exhibit A of the Caraballo settlement agreement lists 26 ``Safe 
Harbor Principles'' that outline the changes to which the parties 
agreed. These principles formed the basis for a new COLA methodology, 
which OPM incorporated into its regulations. In developing these 
regulations, OPM consulted with the Survey Implementation Committee 
(SIC), which was established under the Caraballo settlement and is 
composed of representatives of the parties in Caraballo. The SIC in 
turn consulted with the Technical Advisory Committee (TAC), which was 
also established under the Caraballo settlement and is composed of 
three economists with expertise in living-cost comparisons. OPM 
published proposed regulations incorporating the new methodology in the 
Federal Register for notice and comment on November 9, 2001, at 66 FR 
56741, and a final rule on May 3, 2002, at 67 FR 22339. The SIC and the 
TAC also worked closely with OPM in preparing for and implementing the 
2002, 2003, and 2004 COLA surveys.

COLA Surveys

    In 2002, OPM surveyed Puerto Rico, the U.S. Virgin Islands, and the 
Washington, DC, area. We published the results of these Caribbean 
surveys in the Federal Register on February 9, 2004, at 69 FR 6020. In 
2003, OPM surveyed Anchorage, Fairbanks, and Juneau, Alaska, and the 
Washington, DC, area. We published the results of these Alaska surveys 
on March 12, 2004, at 69 FR 12002. In 2004, OPM surveyed Honolulu 
County, Kailua Kona, Hilo, Kauai County, Maui County, Guam, and the 
Washington, DC, area. We published the results of these Pacific surveys 
on August 4, 2005, at 70 FR 44989. OPM also published the results of 
new shelter (rent) price analyses for the 2002 Caribbean surveys and 
the 2003 Alaska surveys in a Federal Register notice dated August 4, 
2005, at 70 FR 44978.

COLA Rate Changes

    As described in the survey reports, OPM compared the results of 
each of the COLA area surveys with the results of the DC area survey to 
derive a living-cost index for each of the COLA areas. We then added 
adjustment factors as provided in 5 CFR 591.227. The final results 
indicate an increase in the COLA rates for the U.S. Virgin Islands and 
Kauai County, Maui County, and Hawaii County, HI; no change in the COLA 
rates for the Rest of the State of Alaska, Guam and the Northern 
Mariana Islands, and Honolulu County, HI; and a reduction in the COLA 
rates for Puerto Rico and Anchorage, Fairbanks, and Juneau, Alaska.
    Table 1 shows the old and new COLA rates and the survey living-cost 
indexes for each area. Under 5 CFR 591.228(c), COLA rate reductions are 
limited to no more than 1 percentage point per year. The living-cost 
indexes OPM previously published at 70 FR 44979 and 70 FR 44990 have 
been amended based on changes we implemented in response to comments we 
received. These changes are described in the section of this notice on 
Hedonic Regressions. OPM is publishing the new living-cost indexes in a 
Federal Register notice that accompanies this final rule.

                                  Table 1.--COLA Rates and Living-Cost Indexes
----------------------------------------------------------------------------------------------------------------
                                                                    Previously
                                                  Old COLA rates     published    Revised living- New COLA rates
             Allowance area/category                 (percent)      living-cost    cost indexes      (percent)
                                                                      indexes
----------------------------------------------------------------------------------------------------------------
Puerto Rico.....................................            11.5          105.10          103.04            10.5
U.S. Virgin Islands.............................            22.5          122.84          122.53            23.0
Anchorage, Alaska...............................            25.0          113.79          113.64            24.0
Fairbanks, Alaska...............................            25.0          115.61          115.62            24.0

[[Page 43898]]

 
Juneau, Alaska..................................            25.0          118.03          118.09            24.0
Rest of the State of Alaska.....................            25.0          136.00          135.84            25.0
Honolulu County, Hawaii.........................            25.0          127.78          125.80            25.0
Hawaii County, Hawaii...........................            16.5          119.11          117.25        17.0 \1\
Kauai County, Hawaii............................           23.25          130.58          127.63            25.0
Maui County, Hawaii.............................           23.75          134.49          131.50            25.0
Guam and the Northern Mariana Islands...........            25.0          127.65          127.40           25.0
----------------------------------------------------------------------------------------------------------------
 The 2004 Pacific survey report indicated a COLA rate increase for Hawaii County, HI, from 16.5 percent to 19
  percent. OPM, however, refined the rental survey hedonic regressions after taking into consideration comments
  received. The refined methodology results in a 17 percent COLA rate for Hawaii County. The refinements OPM
  adopted pursuant to comments are discussed in the section on Hedonic Regressions.

    OPM published a proposed rule in the Federal Register on August 4, 
2005, at 70 FR 44976, inviting comments on the COLA rate changes. 
Approximately 2,400 commenters responded to the proposed rule. Most of 
the comments were from Federal employees or unions representing Federal 
employees. Many of the commenters expressed opposition, without further 
comment, to the planned COLA rate reductions. Other commenters 
addressed specific issues and concerns with the COLA surveys. One 
agency and two commenters concurred with the rate changes. We summarize 
and evaluate the substantive comments in the ``Discussion of Comments'' 
section that follows.

Discussion of Comments

Rising Living Costs

    Many of the commenters said OPM should not reduce COLA rates 
because their living costs were increasing. A number of commenters 
provided or referred to publications showing rising costs in their COLA 
area. By law, OPM must compare costs in the COLA areas with costs in 
the Washington, DC, area and adjust COLA rates according to the 
relative difference. Therefore, if living costs rise faster in the COLA 
area than in the DC area, we increase the COLA rate subject to the 
statutory maximum. If living costs rise faster in the DC area than in 
the COLA area, we reduce the COLA rate, but by no more than 1 
percentage point per year, as provided by 5 CFR 591.228(c).
    Numerous commenters noted that certain costs have increased since 
OPM conducted the survey in their COLA area and that the survey data 
were outdated. Many commenters requested that OPM survey again. They 
cited the cost of gasoline, housing, utilities, grocery items, medical 
needs, various fees and taxes, and other items. OPM recognizes that 
prices for various items will increase in the COLA areas and/or the DC 
area between surveys. We collect prices in each survey area every 3 
years on a rotating basis according to a schedule agreed upon by the 
parties in the Caraballo settlement. Beginning with the publication of 
the 2005 Caribbean and DC COLA survey results, OPM will adjust price 
indexes for areas not surveyed based on the relative change in the 
Consumer Price Index (CPI) for the COLA area compared with the CPI for 
the Washington, DC, area. Pursuant to the Caraballo settlement, we are 
not implementing CPI adjustments at this time.
    Several commenters said their current living costs are considerably 
higher than living costs at their previous duty station in the lower 48 
States. By law, OPM must compare the cost of living in the COLA areas 
with the cost of living in the Washington, DC, area.

Consumer Goods

    Several commenters noted that some goods available in the 
contiguous States are not available in their COLA area. Other 
commenters said certain items, such as water softeners, are necessary 
in their COLA area, but not in the DC area. Two commenters said they 
pay for certain services, such as garbage collection, that are covered 
by taxes in some other areas. Issues such as these were discussed 
extensively during the COLA litigation. As a result of the Caraballo 
settlement and as provided by 5 CFR 591.227, OPM adds an adjustment 
factor to the price index for each COLA area to reflect differences in 
need, access to and availability of goods and services, and quality of 
life in the COLA area relative to the DC area. With regard to the above 
comments, we note that several water softener companies do business in 
the DC area and assume this reflects a need for some DC area residents 
to have water softeners. We also note that residents in several 
communities in the DC area pay directly for garbage collection.
    One commenter compared prices for a number of items at a department 
store in Puerto Rico with mainland prices listed on the store's 
Internet site, showing the prices for these items in Puerto Rico to be 
higher. The same commenter also remarked on the high cost of vehicles, 
including shipping, in Puerto Rico. Other commenters also noted that 
many consumer goods must be shipped to COLA areas at high cost. In each 
of the annual surveys, OPM contacted over 900 outlets and collected 
more than 4,600 prices on over 240 items representing typical consumer 
purchases. We surveyed the final cost to the consumer of services or 
items, including automobiles. The final cost includes any overhead, 
transportation and shipping costs, taxes, competition, and other price 
influences. Additionally, OPM surveyed catalog prices for a number of 
items and included in the price the costs for shipping, sales tax, and 
excise tax, which are often higher in the COLA area relative to the 
Washington, DC, area.

Inequity Among Areas

    Many commenters claimed the COLA rate reductions in Puerto Rico 
were discriminatory but did not elaborate. Some commenters, however, 
noted that Puerto Rico historically has had the lowest COLA rate of all 
of the COLA areas.
    OPM conducts COLA surveys using the same methodology in all areas. 
For many years, OPM's surveys have indicated that COLA rates should be 
lower in several COLA areas. However, litigation and legislation barred 
OPM from implementing COLA rate reductions. The bars have now expired, 
and we are implementing rate reductions in certain areas. In the 
future, it is possible that there may be more differentiation among 
COLA rates than there is today.
    Several commenters questioned the data collectors' familiarity with 
Puerto

[[Page 43899]]

Rico and knowledge of the Spanish language. OPM used data collectors in 
Puerto Rico who spoke Spanish and had formerly resided in Puerto Rico. 
In addition, the two non-rental OPM data collectors were accompanied by 
observers from the Puerto Rico COLA Advisory Committee (CAC), which is 
composed of current Federal employees who live in Puerto Rico and speak 
Spanish.
    Before each survey, OPM establishes a CAC in each of the survey 
areas. As described in 5 CFR 591.243, each CAC is composed of 
approximately 12 agency and employee representatives from the survey 
area and two representatives from OPM. To help OPM prepare for the COLA 
surveys, OPM and the CACs held 3-day meetings in each area to be 
surveyed to plan the COLA surveys. During the survey, the CAC members 
assisted OPM staff in collecting non-rental data, and after the survey 
the CAC members had the opportunity to review all of the survey 
results, including the results of the rental survey. Although CAC 
members helped plan the rental survey and had the opportunity to review 
the rental survey results in detail, CAC members did not participate in 
the rental data collection as observers.
    A local union in Puerto Rico stated that in the rental survey, OPM 
treated Puerto Rico COLA employees in a disparate fashion because of 
national origin and without regard to unique linguistic and cultural 
differences. The union cited misspellings in the rental data as 
evidence that the data collectors encountered a serious language 
barrier. The OPM contractor that surveyed rental properties in Puerto 
Rico also employed Spanish-speaking data collectors, some of whom were 
or are residents of Puerto Rico. Although there may have been 
misspellings in business names and street addresses, we believe the 
overall quality of the rental data was good and reflected the COLA 
survey's specifications for rental prices in Puerto Rico.

Rental Surveys

    OPM also received from the local union in Puerto Rico extensive 
comments on the Caribbean and DC area rental surveys. A large number of 
commenters wrote in support of the union's comments, made similar 
comments, or addressed other issues relating to the rental surveys. 
Many of the comments addressed 2005 survey issues based on pre-
publication rental data OPM provided to the Puerto Rico CAC for review. 
Because we have not yet published the 2005 survey results and because 
the Puerto Rico rate reduction is based on the 2002 Caribbean survey 
results, we are responding in this final rule to the comments as they 
apply to the 2002 survey.
    The union and other commenters asserted that the rental survey did 
not accurately reflect the areas or the types of housing units where 
Federal employees live in Puerto Rico, was conducted in a careless and 
negligent manner, and was conducted differently in Puerto Rico than in 
the DC area and the U.S. Virgin Islands. Other commenters in Puerto 
Rico charged that the rental survey was inaccurate, unfair, and 
discriminatory, and many challenged the veracity, reliability, and 
adequateness of the rental data.
    OPM hired an experienced contractor to collect rental data in each 
area, and the data were collected in essentially the same manner in all 
areas. In Puerto Rico, we directed the contractor to collect in the San 
Juan/Caguas area and in areas east of San Juan, including the Roosevelt 
Roads area. We selected these areas based on the results of the 1992/93 
Federal Employee Housing and Living Patterns Survey, which provided 
information on where Federal employees live. As described in the survey 
report, OPM collected over 80 housing characteristics on more than 400 
rental observations throughout this area of Puerto Rico and over 900 
observations in the Washington, DC, area. The housing characteristics 
were described in Appendix 4 of the survey report (69 FR 6047).
    Pursuant to the Caraballo settlement, OPM used hedonic regression 
analysis, which is a type of statistical analysis, to compare rents in 
the COLA areas with rents in the DC area. The use of hedonic 
regressions allows OPM to hold quantity and quality of housing constant 
to make rental rate comparisons. The hedonic regressions are described 
in the survey notice at 69 FR 6029 and Appendix 5 at 69 FR 6048. (As 
described in the notice of August 4, 2005, at 70 FR 44798, OPM revised 
these hedonic regressions and published new rent indexes.) Therefore, 
we believe the rental surveys and analyses were conducted in a fair and 
professional manner in all of the COLA areas.
    The union also noted that a high percentage of the Puerto Rico 
observations were gathered from ``drive by'' listing sources. The 
contractor collects information from five types of sources: local 
newspaper/publication, Internet, agent/broker, drive-by/sign posted, 
and other. The contractor collected data from all types except 
``other'' in both Puerto Rico and the DC area, but the distribution of 
observations by listing source type varied by area. To determine 
whether listing source influenced rental rates, OPM added listing 
source as a variable in the hedonic regression analysis. We found that 
the variable was statistically significant, but that it raised the 
standard error of the survey area parameter estimates. Therefore, we 
are not using listing source as a variable in the final hedonic 
regression. (See the section of this notice on Hedonic Regressions.)
    The union also stated that a considerable percentage of the rental 
observations were from individuals who refused to provide some or all 
self-identifying information (i.e., the individual's name and/or his/
her home or business address). Approximately 30 percent of the rental 
observations in Puerto Rico were from such individuals, and 
approximately 4 percent of the observations in the DC area were from 
such individuals. As with listing sources, OPM added self-
identification refusal as a variable to the hedonic regression 
analysis. We found that self-identification refusal was not a 
statistically significant variable and are not using it in the final 
hedonic regression.
    The union and a number of other commenters believe OPM should have 
allowed observers to accompany the contractor on the rental surveys in 
Puerto Rico. The contract did not provide for rental data collection 
observers to accompany contractor data collectors, and we determined 
that a contract modification to allow observers would increase contract 
costs significantly. We are currently exploring with the contractor how 
rental data collection observers might be involved in future COLA 
surveys.
    Various commenters noted expenses that affect the cost of shelter 
in their area. A number of commenters said employees in Puerto Rico pay 
more for living in gated communities or incur costs for security 
systems because of the high crime rate. One commenter remarked on 
higher mortgage rates in Puerto Rico. A commenter from Alaska noted 
that some properties require water wells. To the extent that these 
necessities influence the rental rate for a property, they are captured 
by the rental survey. The rental survey also captures any separate 
security fees added to rents.
    One commenter asked why the COLA survey did not compare homeowner 
data. Under the Caraballo settlement, the parties agreed to adopt a 
rental equivalence approach similar to the one the Bureau of Labor 
Statistics uses for the Consumer Price Index. Rental

[[Page 43900]]

equivalence compares the shelter value (rental value) of owned homes, 
rather than total owner costs, because the latter are influenced by the 
investment value of the home (i.e., what homeowners hope to realize as 
a profit when they sell their homes). As a rule, living-cost surveys do 
not compare how consumers invest their money.
    The same commenter said one of the apartment units surveyed in the 
2003 Juneau survey was listed as 960 square feet, but she said the 
property was actually 690 square feet. OPM asked the contractor to 
verify the information. The contractor found that the units in this 
apartment complex varied in size and that some are 690 square feet, 
while others are 960 square feet. The unit in the Juneau rental 
database was correctly identified at 960 square feet.
    The same commenter also objected to assigning all two-bedroom 
apartments to the same class. For the purposes of drawing the rental 
survey sample, OPM classifies rental units by location into six broad 
categories. The six categories are as follows: Class A--four bedroom, 
single family unit not to exceed 3200 square feet; Class B--three 
bedroom, single family unit not to exceed 2600 square feet; Class C--
two bedroom, single family unit not to exceed 2200 square feet; Class 
D--three bedroom apartment unit not to exceed 2000 square feet; Class 
E--two bedroom apartment unit not to exceed 1800 square feet; and Class 
F--one bedroom apartment unit not to exceed 1400 square feet. OPM uses 
this information only to draw the survey sample. During the survey, OPM 
obtains information on more than 80 housing characteristics, including 
type, actual square footage, number of bedrooms and bathrooms, and 
other characteristics of each rental unit. Instead of comparing houses 
within each of the six broad classes, we use the detailed 
characteristics of each rental unit and hedonic regressions as 
described below to hold these characteristics constant between the COLA 
and Washington, DC, area to make rental price comparisons.
    One commenter asserted that OPM's contractor determined most 
square-footage measurements without visiting the property. OPM's 
contractor visited every property and used, where appropriate, special 
equipment to estimate the square footage of living space. The 
contractor also used living-space information provided by brokers, 
realtors, owners, property tax records, and other reliable sources.

Hedonic Regressions

    As described in the previous section, OPM received several comments 
concerning the rental survey, some of which led us to revisit the 
hedonic regression analyses we use to compute rental indexes. As 
described at 69 FR 6029 and Appendix 1 at 70 FR 44979, OPM used hedonic 
regression analysis, which is a type of multiple linear regression 
analysis, to compare rents in the COLA areas with rents in the DC area. 
Multiple linear regression is used to determine how the dependent 
variable (in this case, rent) is influenced by one or more independent 
variables (in this case, the characteristics of the rental unit 
including some aspects of the neighborhood). As is common in this type 
of analysis and as was done in the research leading to the Caraballo 
settlement, OPM used semi-logarithmic regressions. The regression 
produces parameter estimates for each independent variable, including 
survey area. A parameter estimate is an estimate of the influence of 
rental characteristics or variables on rent of a dwelling unit. 
Variables may be continuous--like square footage, number of bedrooms, 
or number of bathrooms--or class variables, like external condition 
(good, fair, etc.), availability of air conditioning (yes, no), or the 
particular COLA survey area in which the rental unit is located. For 
example, ``Puerto Rico,'' ``St. Croix,'' and ``St. Thomas/St. John'' 
are the Caribbean COLA survey area class variables for which parameter 
estimates are computed.
    COLA survey area parameter estimates are of greatest interest in 
the COLA rental model because, once converted, they become the survey 
area rent index holding all of the other rental characteristics in the 
regression constant. In other words, the exponent of the survey area 
parameter estimate (i.e., after the estimate is converted from natural 
logarithms with a correction for a slight bias caused by the use of 
logarithms) multiplied by 100 is the survey area's rent index. This 
index reflects the difference in rents for the COLA survey area 
relative to the Washington, DC, area, while (in effect) holding other 
significant housing characteristics constant.
    To select the variables to use in the model, OPM adopted a 
methodology developed by the TAC and OPM, in consultation with the SIC. 
The methodology OPM used to produce the rent indexes published in the 
notices (70 FR 44978 and 70 FR 44989) that accompanied the proposed 
rule was an objective, multi-step process by which OPM eliminated 
variables that were not statistically significant. After reviewing the 
results of hedonic regressions OPM performed in response to comments we 
received on the rental survey, the TAC recommended that we refine the 
methodology to also eliminate variables that decreased the precision of 
the rent index. Therefore, OPM modified the variable selection process 
to eliminate variables that are not statistically significant and/or 
decrease the precision of the rent index. The refined methodology 
produces an improved hedonic regression model with somewhat different 
rent indexes than those shown in the survey notices. (See notice that 
accompanies this rule.)
    A commenter from Puerto Rico supported in general the use of 
hedonic techniques, but was critical of the variables in the OPM 
regression model. The commenter noted a number of characteristics he 
thought OPM should have included. OPM collected most of the 
characteristics the commenter suggested. (See Appendix 4 at 69 FR 
6047.) We then processed the characteristics using the methodology 
described above and included many of them (e.g., square footage, number 
of bedrooms and bathrooms) in the final hedonic regression analysis. 
Therefore, although OPM collects more than 80 rental unit 
characteristics, the multi-step process described above produces a 
hedonic regression model with fewer characteristics, i.e., those that 
are statistically most important and increase precision in terms of the 
rent indexes.
    OPM does not collect some of the variables (e.g., occupancy ratio) 
noted by the commenter. These variables are often included in studies 
related to the appraised value of properties, particularly apartment 
complexes, because they might be important to a prospective commercial 
buyer. We do not collect information such as occupancy ratios, however, 
because they are not important for renters of single units within a 
complex.
    The commenter noted that location is an important variable and 
recommended collecting information on neighborhoods and ``distance to 
major employment centers.'' OPM added distance as a variable in the 
hedonic regression analysis by computing the distance from each rental 
observation to the major Federal building or intersection in each 
survey area. We then treated distance as any other variable in the 
model and examined its significance and its impact. As it turned out, 
distance did not enter as a variable in any of the final models because 
it either was not significant or decreased the precision of the rent 
index.
    With regard to neighborhoods, as shown in Appendix 4 at 69 FR 6047, 
OPM already collects information that reflects the quality of the 
neighborhood.

[[Page 43901]]

We use information from the Bureau of the Census to introduce 
additional variables to the hedonic regressions that may reflect the 
quality of the neighborhood. To do this, we identify the census tract 
within which each rental observation is found and then add variables, 
such as median income, percent school age persons, and percent of 
people in the area with B.A. degrees or higher, to the hedonic 
regressions. We process these characteristics using the methodology 
described above and those that are statistically significant and 
increase precision are used in the final hedonic regression analysis.
    OPM was not able to add census tract data to the 2002 survey. We 
did not have longitude and latitude coordinates for the 2002 rental 
observations, and we are not aware of any software product that could 
provide this information using Puerto Rico and U.S. Virgin Islands 
addresses. We note, however, that we collected longitude and latitude 
information in the 2005 Caribbean survey, and we anticipate using 
census tract data in the analysis of the rental survey results.
    The commenter described a type of logit model that appraisers find 
useful in distinguishing ``atypical'' apartment complexes--possibly an 
important tool in the appraisal field. Further, the logit model is 
useful when the dependent variable is limited in range (e.g., the 
probability of buying an apartment building of given characteristics) 
but is not appropriate for a continuous variable, like rents for 
similar accommodation across cities, which is the goal of OPM analysis.
    The commenter said OPM did not say which statistical package it 
used and did not describe its analyses. As stated at 70 FR 44978, OPM 
used SAS, which is a common proprietary statistical package. Appendices 
1 and 2 at 70 FR 44979 show the details of the regression models OPM 
used and the results of the regressions.
    The commenter also said OPM should provide the statistical 
procedures used for the hedonic regressions to the CACs for review. OPM 
provides the CACs with COLA survey materials that explain regression 
analysis, contain graphs and charts, and provide the same details about 
the rental survey and hedonic regressions OPM publishes in its Federal 
Register notices. In addition, OPM staff meets with the CACs to explain 
the procedures used, go over the hedonic regression results, and answer 
questions.

Locality Pay and Retirement

    Numerous commenters said Federal employees in the COLA areas should 
receive locality pay. Several requested that OPM replace the COLA rate 
with locality pay or take DC area locality pay into consideration when 
setting COLA rates. Two commenters noted that the Rest of U.S. (RUS) 
locality pay rate is higher than the COLA rate in Puerto Rico. One 
commenter said OPM should set the lowest COLA rate to be equivalent to 
the RUS locality pay rate. Several commenters noted that locality pay 
is included in base pay for retirement purposes, while COLAs are not 
included. One commenter said that not considering COLAs in retirement 
calculations creates a disincentive to retire in a COLA area. Another 
commenter said that employees in the COLA areas have to save more for 
retirement.
    The Federal Employees Pay Comparability Act of 1990 (FEPCA) 
authorizes locality pay only for Federal employees in the contiguous 48 
States and Washington, DC. OPM cannot consider DC area locality pay or 
set COLA rates at certain locality pay levels because doing so would be 
tantamount to implementing locality pay outside the 48 States. 
Additionally, OPM cannot credit COLAs in the retirement calculation 
because 5 U.S.C. 8331(3) and 8401(4) exclude allowances from base pay 
for Federal retirement purposes. Changes in law would be required to 
extend locality payments to Federal employees in the COLA areas or to 
include COLAs in base pay for Federal retirement purposes.

Recruitment and Retention

    Many commenters believe COLA reductions would cause recruitment and 
retention problems. OPM is concerned about the Government's ability to 
recruit and retain a well-qualified workforce and notes that the 
Government has several pay authorities that are available to address 
recruitment and retention problems. Among these are special salary 
rates and recruitment, retention, and relocation incentives.

Financial Effect

    Many commenters were concerned about the impact of COLA reductions 
on personal financial commitments, such as mortgages, and other 
financial obligations. Several commenters stated that the reductions 
would have an adverse effect on the local economy of the COLA area. As 
noted above, OPM's authority to set COLA rates is established in 5 
U.S.C. 5941. However, our regulations provide that COLA rates may be 
reduced by no more than 1 percentage point in any 12-month period, 
which serves to minimize the financial impact of COLA rate reductions.

Utility Costs

    Several commenters remarked that water, electricity, and other 
utility costs are high in Alaska and Puerto Rico. One commenter noted 
that increasing energy costs also affect costs for shelter, 
transportation, and consumer goods. OPM surveyed utility costs and 
included these costs in the price comparisons. We also surveyed costs 
for shelter, transportation, and consumer goods and services. The 
prices of goods and services include any energy and/or local 
transportation costs associated with making these items available for 
sale.
    Some commenters reported that water utility prices have or are 
going to increase significantly in Puerto Rico. To the extent that such 
increases occur, they will be reflected in the results of the 2005 
survey or in subsequent surveys and/or adjustments.

Transportation

    Several commenters noted the cost of long distance travel from the 
COLA areas to areas in the continental United States. The commenters 
requested that OPM consider time, distance, and excessive travel 
expenses in setting COLA rates. Two other commenters noted higher air 
transportation costs in Juneau because of the lack of airline 
competition.
    The COLA methodology takes travel expenses into account in two 
ways. First, OPM compares the cost of air travel from the various COLA 
areas to common destinations in the contiguous States with the cost of 
air travel from the DC area to those same destinations. This would 
capture any higher ticket prices that result from reduced competition. 
Second, as provided in 5 CFR 591.227, OPM adds to the overall price 
index for the COLA area an adjustment factor that reflects differences 
in need, access to and availability of goods and services, and quality 
of life in the COLA area relative to the DC area. This adjustment 
factor is designed to address such considerations as the difficulty of 
traveling long distances.

Medical Services

    Several commenters believe the survey underestimated the cost and 
restricted availability of medical services in Alaska. They also noted 
that doctor visits and dental care are more expensive in the COLA 
areas. One commenter said none of the Federal health benefit plans in 
Juneau offer supplemental dental coverage. Another commenter felt that 
medical care in

[[Page 43902]]

Juneau was limited, resulting in higher health care costs and inferior 
health care. Several commenters said there was a need for costly travel 
outside the area to obtain some medical services.
    OPM surveys the prices of several medical services (including 
dental services) and items in each COLA area and in the DC area. The 
medical services index reflects any relatively higher local prices. The 
availability of medical services is not something OPM prices or 
quantifies. Instead, it is part of the adjustment factor OPM adds to 
the price index to reflect differences in need, access to, and 
availability of goods and services, and quality of life in the COLA 
areas relative to the Washington, DC, area.
    One commenter noted that there were no Health Maintenance 
Organizations (HMOs) in Juneau. As described at 69 FR 12005, OPM 
compared average health insurance premium costs in the COLA area with 
average health insurance premium costs in the DC area. Therefore, the 
health insurance premium index reflects higher local costs to the 
extent that an area has only higher cost plans available (i.e., to the 
extent HMOs are not available).

Quality of Life

    A number of commenters stated that the COLA reductions would affect 
their quality of life. One commenter from Alaska said the COLA is an 
incentive that helps with the isolation, extreme climate, support 
issues, and darkness. As noted under the Transportation section, OPM 
adds adjustment points in part to compensate for differences in quality 
of life.

Taxes

    Several commenters mentioned increased sales, excise, and local 
taxes in Puerto Rico. Another commenter noted that Federal employees in 
North Pole, Alaska, must pay a sales tax. OPM adds sales tax and, where 
applicable, excise and other taxes to the prices it collects. In Puerto 
Rico, excise taxes paid by importers and distributors are part of the 
price for the item. In the case of catalog items, OPM adds such taxes 
as applicable. To the extent any recent tax increases in Puerto Rico 
have occurred, they will be reflected in the results of the 2005 survey 
or in subsequent surveys and/or adjustments.

Local Conditions

    Several commenters from Puerto Rico noted additional costs faced by 
Federal employees on the island because of hurricanes and blackouts. 
Among the costs mentioned were generators, special water tanks, storm 
shutters, bottled water, road damage, electrical equipment repair, and 
higher property insurance. OPM discusses property insurance under the 
heading Insurance. As noted under the Transportation section, OPM adds 
adjustment points pursuant to 5 CFR 591.227 in part to compensate for 
differences in quality of life.
    A number of commenters also remarked on weather conditions in 
Alaska. They noted additional costs, such as four-wheel drive, studded 
tires, winter clothes, high electric and heating bills, and vehicle 
maintenance, because of the long winters, icy roads, and temperatures 
that sometimes extend to 40 degrees or more below zero. As described in 
Appendix 3 at 69 FR 12027, OPM priced in Alaska a four-wheel drive 
vehicle with an engine block heater and regular and studded tires. We 
also priced parkas, boots, and other cold weather items. The utility 
model we use reflects Alaska's higher home energy costs.

Insurance

    Several commenters noted high property insurance rates as a result 
of escalating housing prices, hurricanes, or property located in flood 
zones. OPM uses a rental equivalence approach to determine shelter 
costs. The rental equivalence approach compares the rental values of 
homes. Home insurance is implicit in these values. Therefore, we do not 
survey any type of homeowner insurance, but we do survey renter 
insurance. In doing so, we include the price of any special riders 
necessary to cover hurricane or typhoon damage.

Education

    Several commenters cited the necessity for placing children in 
private schools in Puerto Rico. The commenters noted language, quality, 
and danger issues with the public schools. One commenter said that the 
quality of Puerto Rico public schools is poor, while the DC area has 
some of the best public schools in the nation. Another commenter said 
employees in Puerto Rico who want to compete for employment 
opportunities in the mainland must place their children in costly 
English language private schools. Several commenters remarked on the 
high cost of private school tuition, school supplies, and various 
school fees in Puerto Rico. Other commenters said the Department of 
Defense school in San Juan is not available to many Federal employees, 
so they must pay for private schools. One commenter from the island of 
Hawaii said school choices and day care facilities there are limited, 
and the only private school for high school students is in Waimea. 
Another commenter said private schools in Juneau are limited, 
increasing costs for those who must use out-of-state private schools.
    OPM surveyed K-12 private education in the COLA and DC areas and 
computed an average tuition price that reflected all grade levels. 
Because not everyone sends children to private school, we made an 
additional special adjustment for K-12 education by applying ``use 
factors.'' These use factors reflect the relative extent to which 
Federal employees make use of private education in the COLA and DC 
areas. OPM described the process used for K-12 private education in the 
Caribbean region at 69 FR 6030, in Alaska at 69 FR 12007, and in the 
Pacific region at 70 FR 44995.
    Two commenters said many Federal employees in Puerto Rico send 
their children to colleges in the continental United States. A 
commenter from Alaska said there were limited colleges and universities 
in Alaska, so many employees send their children to colleges in the 
lower 48 States and incur extra costs for non-resident tuition and 
transportation. Another commenter said tuition at the University of 
Puerto Rico will be increasing. Two commenters said Federal employees 
and their families are not eligible for student grants in Puerto Rico.
    OPM does not measure the price of college and university education 
because where employees send their dependents to school is often a 
matter of personal preference. For example, many Federal employees in 
the Washington, DC, area send their children to colleges and 
universities outside the DC area. To the extent OPM leaves an item, 
such as college education, out of the COLA model, the effect is as if 
OPM included it in the model at the overall price index for the area. 
Therefore, if prices are generally higher in a COLA area relative to 
the DC area, the implicit assumption is that college and university 
prices are higher to the same extent. Any additional costs would be 
reflected in the adjustment points added pursuant to 5 CFR 591.227.

Geographic Coverage

    One commenter said there should be separate COLA rates for the east 
(Kona) and west (Hilo) sides of the island of Hawaii because prices are 
not equal. OPM does not plan to split Hawaii County into separate areas 
at this time, but OPM may reconsider that decision after additional 
surveys have been completed using the methodology

[[Page 43903]]

adopted pursuant to the Caraballo settlement.
    One commenter requested that OPM consider Eielson Air Force Base as 
part of the Rest of Alaska COLA area. The commenter noted that Eielson 
is 26 miles from Fairbanks and that it is dangerous to drive that 
distance for groceries, the hospital, or the airport during the winter 
months. The commenter suggested changing the distance parameter for 
Fairbanks to 20 miles. As stated in 5 CFR 591.206(b), the head of a 
department or agency must submit a request to OPM to initiate any 
reconsideration of the definition of a COLA area. We note that North 
Pole, AK, is only 8 miles from Eielson, and that North Pole is part of 
the Fairbanks survey area. OPM surveys a supermarket and other 
businesses in North Pole and includes these in the calculation of the 
Fairbanks living-cost index. We do not believe it would be appropriate 
to change the definition of the Fairbanks, AK, COLA area.
    Another commenter said the cost of living in Girdwood, AK, which is 
40 miles outside of Anchorage, is significantly higher than in 
Anchorage. The commenter asked that OPM consider the effect of the COLA 
reduction on employees living and working in outlying communities of 
Anchorage. As noted above, the head of a department or agency must 
submit a request to OPM to initiate any reconsideration of the 
definition of a COLA area.
    One commenter noted that 5 U.S.C. 5941 states that COLAs are based 
on ``living costs substantially higher than in the District of 
Columbia.'' The commenter said the area OPM uses for comparison 
includes areas in addition to the District of Columbia. The commenter 
referred also to Executive Order 10000, which uses the term 
``Washington, DC, area'' but precedes it with the phrase ``subject to 
applicable law.'' The commenter requests that OPM survey only the 
District of Columbia. The President directs OPM in Executive Order 
10000 to designate nonforeign areas at locations in which living costs 
are substantially higher than in the Washington, DC, area and set COLA 
rates for such areas based on these higher living costs. OPM does not 
plan to limit the scope of the Washington, DC, area survey to the 
District of Columbia.

Survey Rates

    An agency and one other commenter requested that OPM address the 
total anticipated COLA rate reductions projected by the area surveys. 
OPM conducts COLA surveys once every 3 years and will adjust COLA rates 
pursuant to 5 CFR 591.224, beginning with the publication of the 
results of the 2005 Caribbean survey. Therefore, we cannot predict what 
COLA rates will be in future years. However, with each survey notice 
OPM does publish final living-cost indexes that can easily be converted 
to hypothetical COLA rates (prior to the 1 percentage point limitation 
on COLA rate reductions) with the application of 5 CFR 591.228(a). This 
essentially involves converting the living-cost index to a percentage 
and rounding the result to the nearest whole percentage point. Table 1 
includes the indexes that can be used for this purpose. However, it 
should be noted that future surveys and adjustments will likely produce 
different results.

Employee Involvement

    Several commenters believe employees from their area were not 
involved in the COLA surveys. As noted above, OPM established and 
worked with local COLA Advisory Committees (CACs) in each survey area.
    One commenter asked who represented Juneau on the CAC. The Juneau 
CAC was composed of representatives from the Juneau Federal Executive 
Association, the Juneau COLA Defense Committee, the National Federation 
of Federal Employees, the Indian Educators Federation-American 
Federation of Teachers, the National Weather Service Employees 
Organization, Professional Airways Systems Specialists, the Department 
of Agriculture, the Department of Commerce, the Department of the 
Interior, the Department of Transportation, and OPM.
    The same commenter wanted to know more about Federal employee input 
into the COLA survey, units of measure, and formulas used in analyzing 
the data. Federal employees serve on both the SIC and the CAC. As 
explained in the Background section, the SIC worked closely with OPM as 
we developed new COLA regulations pursuant to the Caraballo settlement. 
The composition of the SIC is described in the Caraballo stipulation 
for settlement, which is available on OPM's Web site at http://www.opm.gov/oca/cola/settlement. The regulations we adopted describe 
the methodology and formulas used to analyze the survey data. These 
regulations can be found on OPM's Web site at http://www.opm.gov/oca/cola/RegsRpts.asp.
    As also explained in this final rule OPM worked closely with the 
CACs to plan, conduct, and review the results of the COLA surveys. The 
CACs are described in OPM regulations at 5 CFR 591.240 to 591.244. 
Additional information about the surveys and analyses used may be found 
in the survey reports, which are on OPM's Web site at http://www.opm.gov/oca/cola/index.asp.

Military COLA

    Several commenters thought the reductions in the proposed 
regulation applied to the military COLA. These reductions apply only to 
the COLA rates paid to certain civilian white-collar Federal employees 
paid under the General Schedule and similar pay plans. Three commenters 
remarked on the discrepancy between the civilian and military COLAs. 
The Department of Defense sets the military COLA using a different 
methodology as authorized under separate law. The methodology for the 
civilian nonforeign area COLA derives from 5 U.S.C. 5941, Executive 
Order 10000, and the Caraballo settlement.

Military Post Privileges

    One commenter said all Federal employees should be allowed to shop 
at military commissaries/exchanges. The Department of Defense operates 
commissaries and exchanges. OPM does not have authority to regulate 
commissary/exchange access.

Federal Wage System Employees

    One employee felt Federal Wage System (FWS) employees should 
receive the nonforeign area COLA. The law that authorizes nonforeign 
area COLAs (5 U.S.C. 5941) allows payment of COLAs to employees whose 
rates of pay are set by statute. When the COLA law was enacted, FWS pay 
was set administratively according to local prevailing rates, rather 
than by statute. Currently, FWS rates of pay are not set by statute, 
and OPM cannot extend COLAs to FWS employees.

Communication of Changes

    Three commenters thought OPM did not properly communicate the COLA 
reductions to Federal employees. The Administrative Procedure Act 
requires agencies to publish regulations in the Federal Register as a 
means of notifying the public of rule changes. In addition to 
publishing the proposed regulation in the Federal Register, OPM 
distributed the regulation to agencies with a notice to be posted on 
employee bulletin boards. OPM also summarized and linked to the 
regulations on its Web site at http://www.opm.gov/fedregis/html/aug05.asp and provided copies to COLA Advisory Committee members in 
each area. As noted above, OPM received

[[Page 43904]]

more than 2,000 responses from COLA area employees during the public 
comment period on the proposed regulations.

Correction

    One commenter pointed out an error in the shelter index for Kauai 
in Appendix 7 of the 2004 Survey Report. This error was made in 
typesetting the survey notice. The ``1'' that precedes the PEG index 
belongs to the previous column, so that the PEG Weight should show 
``89.01'' and the PEG Index should show ``118.21.'' Because it was a 
typesetting error, it does not affect OPM's calculations for Kauai.

Executive Order 12866, Regulatory Review

    This rule has been reviewed by the Office of Management and Budget 
in accordance with Executive Order 12866.

Regulatory Flexibility Act

    I certify that this regulation will not have a significant economic 
impact on a substantial number of small entities because it will affect 
only Federal agencies and employees.

List of Subjects in 5 CFR Part 591

    Government employees, Travel and transportation expenses, Wages.

Office of Personnel Management.
Linda M. Springer,
Director.

0
Accordingly, the Office of Personnel Management amends subpart B of 5 
CFR part 591 as follows:

PART 591--ALLOWANCES AND DIFFERENTIALS

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

0
1. The authority citation for subpart B of 5 CFR part 591 continues to 
read as follows:

    Authority: 5 U.S.C. 5941; E.O. 10000, 3 CFR, 1943-1948 Comp., p. 
792; and E.O. 12510, 3 CFR, 1985 Comp., p. 338.


0
2. Revise appendix A of subpart B to read as follows:

Appendix A to Subpart B of Part 591--Places and Rates at Which 
Allowances are Paid

    This appendix lists the places approved for a cost-of-living 
allowance and shows the authorized allowance rate for each area. The 
allowance percentage rate shown is paid as a percentage of an 
employee's rate of basic pay. The rates are subject to change based 
on the results of future surveys.

------------------------------------------------------------------------
                                                              Allowance
                    Geographic coverage                          Rate
                                                              (percent)
------------------------------------------------------------------------
State of Alaska:
    City of Anchorage and 80-kilometer (50-mile) radius by          24.0
     road..................................................
    City of Fairbanks and 80-kilometer (50-mile) radius by          24.0
     road..................................................
    City of Juneau and 80-kilometer (50-mile) radius by             24.0
     road..................................................
    Rest of the State......................................         25.0
State of Hawaii:
    City and County of Honolulu............................         25.0
    Hawaii County, Hawaii..................................         17.0
    County of Kauai........................................         25.0
    County of Maui and County of Kalawao...................         25.0
Territory of Guam and Commonwealth of the Northern Mariana          25.0
 Islands...................................................
Commonwealth of Puerto Rico................................         10.5
U.S. Virgin Islands........................................         23.0
------------------------------------------------------------------------

[FR Doc. 06-6624 Filed 8-1-06; 8:45 am]
BILLING CODE 6325-39-P