Standard for the Flammability of Residential Upholstered Furniture

Federal RegisterMar 4, 2008

Ask Donna

What actually matters in this document.

Text

CONSUMER PRODUCT SAFETY COMMISSION

16 CFR Part 1634

Standard for the Flammability of Residential Upholstered Furniture

AGENCY:

Consumer Product Safety Commission.

ACTION:

Notice of proposed rulemaking.

SUMMARY:

The United States Consumer Product Safety Commission (“Commission” or “CPSC”) is proposing flammability standards for residential upholstered furniture under the Flammable Fabrics Act (“FFA”). The proposal would establish performance requirements and certification and labeling requirements for upholstered furniture. Manufacturers of upholstered furniture would choose one of two possible methods of compliance: They could use cover materials that are sufficiently smolder resistant to meet a cigarette ignition performance test; or they could place fire barriers that meet smoldering and open flame resistance tests between the cover fabric and interior filling materials. Manufacturers of upholstered furniture would be required to certify compliance with the standard and to comply with certain recordkeeping requirements as specified in the proposal.

DATES:

Comments in response to this document must be received by the Commission not later than May 19, 2008.

Comments on elements of the proposed rule that, if issued in final form would constitute collection of information requirements under the Paperwork Reduction Act, may be filed with the Office of Management and Budget (“OMB”) and with the Commission. Comments will be received by OMB until May 5, 2008.

ADDRESSES:

Comments should be filed by e-mail to

cpsc-os@cpsc.gov.

Comments also may be filed by telefacsimile to (301) 504-0127 or mailed, preferably in five copies, to the Office of the Secretary, Consumer Product Safety Commission, 4330 East West Highway, Bethesda, MD 20814, or delivered to the Office of the Secretary, Consumer Product Safety Commission, Room 502, 4330 East-West Highway, Bethesda, Maryland; telephone (301) 504-7530. Comments should be captioned “Upholstered Furniture NPR.”

Comments to OMB should be directed to the Desk Officer for the Consumer Product Safety Commission, Office of Information and Regulatory Affairs, OMB, Washington, DC 20503. The Commission asks commenters to provide copies of such comments to the Commission's Office of the Secretary, with a caption or cover letter identifying the materials as comments submitted to OMB on the proposed collection of information requirements for the proposed upholstered furniture flammability standard.

The public may also request an opportunity to present comments orally. Such requests should be submitted to the Office of the Secretary of the Commission by e-mail, mail, fax or in person at the addresses or phone numbers listed above for the CPSC.

FOR FURTHER INFORMATION CONTACT:

Dale R. Ray, Project Manager, Directorate for Economic Analysis, Consumer Product Safety Commission, 4330 East West Highway, Bethesda, MD 20814; telephone (301) 504-7704.

SUPPLEMENTARY INFORMATION:

A. Background

Regulatory/technical activity

. In 1993 the National Association of State Fire Marshals (“NASFM”) petitioned the Commission to issue regulations under the FFA addressing upholstered furniture fire risks. NASFM requested that the Commission adopt three existing state of California standards.

The Commission granted the petition in part, and issued an advance notice of proposed rulemaking (“ANPR”) on June 15, 1994 on the specific risk of small open flame-ignited fires. 59 FR 30,735 (1994). The Commission denied the petition with respect to large open flame-ignited fires, and deferred action on the petition with respect to cigarette-ignited fires pending a CPSC staff evaluation of: (1) The level of voluntary conformance to existing voluntary industry guidelines, and (2) the overall level of cigarette ignition resistance among products on the market.

Following issuance of the 1994 ANPR, CPSC staff developed a draft performance standard and a test method to evaluate the small open flame performance of upholstered furniture. In October 1997, the staff forwarded a briefing package to the Commission concluding that a small open flame standard was feasible and could effectively reduce the risk to consumers, including both small open flame and cigarette ignitions. The staff recommended that the Commission defer action until the agency could gather additional scientific information to ensure that flame retardant (“FR”) upholstery fabric treatments that manufacturers might use would not result in adverse health effects. The staff recommended that the Commission defer action on the cigarette ignition portion of the 1993 NASFM petition pending a decision on open flame ignition. On October 5, 2001, NASFM withdrew the portion of the petition seeking Commission action with respect to cigarette-ignited fires.

In July of 2003 the CPSC staff recommended that the Commission issue an ANPR to expand the upholstered furniture proceeding to address ignition of upholstered furniture by both small open flames and by smoldering cigarettes. The Commission accepted the staff's recommendation, and the ANPR was published on October 23, 2003. 68 FR 60,619. The 2003 ANPR sought comment on issues relating to the kinds of standard provisions that might best address the upholstered furniture fire risk in its entirety.

The Commission received 13 written comments during the 60-day formal comment period following publication of the ANPR. Interested parties subsequently provided additional written submissions in the form of letters, position statements or presentations of technical data at meetings. A detailed discussion of significant comments received is provided in Section G of this preamble. In October 2004, the staff held a public meeting to present the direction of what would become the staff's 2005 draft standard. The staff analyzed comments received at that meeting as well. The proposed standard takes account of that analysis. Staff received comments on its 2005 draft standard, continued its research and analysis and developed a revised, 2007 draft proposal that focused primarily on preventing smoldering ignitions and reducing the need for flame retardant chemicals.

This notice presents the 2007 draft as the Commission's proposed standard.

The Commission staff briefing package discussing this proposal, Briefing Package: Regulatory Alternatives for Upholstered Furniture Flammability, November 2007, (the “Staff Briefing Package”) is available on the Internet at:

http://www.cpsc.gov/library/foia/foia08/brief/briefing.html.

Copies may also be requested from the Commission's Office of the Secretary at the address shown above.

Acting Chairman Nancy Nord and Commissioner Thomas H. Moore issued statements which are available from the Commission's Office of the Secretary (see

Addresses

section of this notice) or from the Commission's Web site,

http://www.cpsc.gov/pr/statements.html

Overview of the proposed standard

. The proposed standard establishes two possible approaches. Upholstered furniture can meet the proposed standard by having either (1) upholstery cover material that complies with the prescribed smoldering ignition

resistance test (referred to as “Type I” furniture) or (2) an interior fire barrier that complies with specified smoldering and small open flame ignition resistance tests (“Type II” furniture). No requirements are prescribed for filling materials. The standard would become effective one year after issued in final form and would apply to upholstered furniture manufactured or imported on or after that date.

The performance tests prescribed in the proposed standard are conducted with the tested material installed in mockups that simulate the intersection of the seating area of an item of upholstered furniture. In addition to the material under test, the mockup is assembled using standardized upholstery test materials as defined in the proposed standard.

Manufacturers (including importers) of upholstered furniture would be required to certify that the article of upholstered furniture complies with the proposed standard and to maintain records demonstrating compliance with the applicable portions of the proposed standard. Upholstered furniture subject to the proposed standard would be required to be labeled with information identifying the manufacturer, the date of manufacture, the item and type of furniture, and a statement certifying that the article complies with applicable requirements of the standard.

B. Statutory Authority

This proceeding is conducted pursuant to Section 4 of the Flammable Fabrics Act (“FFA”), which authorizes the Commission to initiate proceedings for a flammability standard when it finds that such a standard is “needed to protect the public against unreasonable risk of the occurrence of fire leading to death or personal injury, or significant property damage.” 15 U.S.C. 1193(a).

Section 4 also sets forth the process by which the Commission may issue a flammability standard. As required in section 4(g), the Commission has issued an ANPR. 68 FR 60629. 15 U.S.C. 1193(g). The Commission has reviewed the comments submitted in response to the ANPR and now is issuing a notice of proposed rulemaking (“NPR”) containing the text of the proposed rule along with alternatives the Commission has considered and a preliminary regulatory analysis. 15 U.S.C. 1193(i). The Commission will consider comments provided in response to the NPR and decide whether to issue a final rule along with a final regulatory analysis.

Id.

1193(j). The FFA also requires that when issuing a standard or regulation the Commission must provide an opportunity for interested persons to present their views orally.

Id.

1193(d).

The Commission cannot issue a final rule unless it makes certain findings and includes these in the regulation. The Commission must find: (1) If an applicable voluntary standard has been adopted and implemented, that compliance with the voluntary standard is not likely to adequately reduce the risk of injury, or compliance with the voluntary standard is not likely to be substantial; (2) that benefits expected from the regulation bear a reasonable relationship to its costs; and (3) that the regulation imposes the least burdensome alternative that would adequately reduce the risk of injury. 15 U.S.C. 1193(j)(2). In addition, the Commission must find that the standard (1) is needed to adequately protect the public against the risk of the occurrence of fire leading to death, injury or significant property damage, (2) is reasonable, technologically practicable, and appropriate, (3) is limited to fabrics, related materials or products which present unreasonable risks, and (4) is stated in objective terms.

Id.

1193(b).

C. The Product

The proposed standard applies to residential upholstered furniture. The proposal specifically requires testing of cover fabrics and, alternatively, barrier materials if they are used as a means of complying with the proposed standard. Upholstered furniture is defined for purposes of the proposed standard to include articles of interior seating furnishing intended for indoor use in a home or other residential occupancy that: (1) Consist in whole or in part of resilient cushioning materials (such as foam, batting, or related materials) covered by fabric or related materials; and (2) are constructed with a contiguous upholstered seat and back or arms. Included within the definition are products that are intended or promoted for indoor residential use for sitting or reclining upon, such as: Chairs, sofas, motion furniture, sleep sofas, home office furniture customarily offered for sale through retailers or otherwise available for residential use, and upholstered furniture intended for use in dormitories or other residential occupancies. Items excluded from the definition are: Furniture, such as patio chairs, intended solely for outdoor use; furniture without contiguous upholstered seating and backs and/or arm surfaces, such as ottomans, pillows or pads that are not sold with the article of furniture; commercial or industrial furniture not offered for sale through retailers or not otherwise available for residential use; furniture intended or sold solely for use in hotels and other short-term lodging and hospitality establishments; futons, flip chairs, the mattress portions of sleep sofas, and non-furniture infant or juvenile products such as walkers, strollers, high chairs or pillows.

Commission staff estimates that the proposed standard would affect more than 1,600 manufacturers and importers of upholstered furniture and the 100-200 textile manufacturers that derive a significant share of their revenues from household furniture fabrics. The staff estimates that the average useful life of upholstered furniture ranges from 15 to 17 years. Assuming that the expected life of a piece of upholstered furniture is about 16 years, the average number of upholstered furniture items in household use during 2002-2004 was about 447 million pieces. Upholstered furniture products and manufacturers are discussed in greater detail in section H,

Preliminary Regulatory Analysis

, of this preamble.

The top four companies accounted for nearly 35 percent of the total value of household upholstered furniture shipments in 2002; the 50 largest companies accounted for about 67 percent. The industry also includes many small companies. The staff estimates that nearly all of the affected firms (over 97 percent) would be classified as small businesses under Small Business Administration guidelines. The staff's initial analysis of the potential impact of the proposed standard on such “small entities” is provided in section I.,

Initial Regulatory Flexibility Analysis

, of this preamble.

As discussed in section D of this preamble, the majority of deaths and injuries resulting from fires involving upholstered furniture were started by smoldering ignition sources (such as cigarettes). The staff's test data show that furniture covered with predominantly cellulosic fabrics (such as cotton and rayon) is much more likely to be involved in cigarette-ignited fires than furniture covered with predominantly thermoplastic fabrics (such as polyester, polyolefin, and nylon). The proposed standard focuses primarily on reducing deaths and injuries from smoldering ignited fires. Staff estimates that about 14 percent of currently-produced furniture items are likely to fail the proposed standard's smoldering ignition test for cover fabrics. These would primarily be items constructed with certain predominantly cellulosic fabrics; staff believes most of these fabrics could be modified to meet the proposed standard. Staff anticipates that most manufacturers are likely to bring these furniture items into

compliance by modifying the physical characteristics of the cover fabrics rather than by using flame retardant (FR) fabric treatments. Alternatively, manufacturers would have the option to meet the proposed standard by using barrier materials that pass open flame and smoldering ignition tests rather than changing the cover fabric.

D. Risk of Injury

Annual estimates of national fires and fire losses involving ignition of upholstered furniture are based on data from the U.S. Fire Administration's National Fire Incident Reporting System (“NFIRS”) and the National Fire Protection Administration's (“NFPA”) annual survey of fire departments.

National fire loss estimates for 2002-2004 indicated that upholstered furniture was the first item to ignite in an average 7,800 residential fires attended by the fire service annually during that period. These fires resulted in an average of 540 deaths, 870 injuries and $250 million in property loss each year.

Of these fires, the staff considers an estimated 3,500 fires, 280 deaths, 500 injuries, and $112 million property loss annually to be addressable by the proposed standard. Addressable here means the incidents were of a type that would be affected by the proposed standard (i.e., a fire that ignited upholstered furniture and that had a smoking material or small open flame heat source). Approximately 90% of estimated deaths, 65% of estimated injuries and 59% of property damage resulted from ignition by smoking materials, almost always cigarettes. The remaining addressable fires were started by small open flame sources. Among the addressable casualties, smoking materials accounted for about 260 deaths and 320 injuries annually. Small open flame fires accounted for about 30 deaths and 170 injuries annually.

Numbers do not add up to totals due to rounding.

E. Other Upholstered Furniture Flammability Standards

1. California Regulatory Activity

California Technical Bulletin 117 (“TB-117”), the mandatory regulation for all upholstered furniture sold in that state, contains both smoldering and small open flame resistance performance requirements. Complying upholstered furniture is generally similar to furniture sold in other states, except that California furniture is typically made with FR resilient foam filling materials. In early 2002, the California Bureau of Home Furnishings and Thermal Insulation (BHFTI) released a draft revision of TB-117. This draft revision contained upgraded performance requirements for small open flame ignition resistance of filling materials, and a cover material test similar to that developed by the Commission staff in its 2001 draft small open flame standard. The TB-117 smoldering resistance provisions were not changed.

The California BHFTI has not proposed amendments to TB-117 to incorporate the 2002 draft revision. The BHFTI's comment on the Commission's October 23, 2003 ANPR expressed support for a uniform national standard. BHFTI recommended that the Commission consider adopting appropriate elements of the 2002 draft revised TB-117 into a proposed Commission rule. The proposed standard contains some requirements similar to provisions of TB-117.

2. United Kingdom Regulations

The U.K. Department of Trade and Industry (“DTI”) enforces the U.K. Furniture and Furnishings (Flammability) Regulations, issued in 1990. These regulations contain smoldering and open flame resistance requirements for residential upholstered furniture based on test methods in British Standard BS 5852. The CPSC proposed standard's fire barrier open flame test uses the apparatus and ignition source from the U.K. regulations.

3. Voluntary Standards Activity

Since the Commission's original ANPR on upholstered furniture was published in 1994, industry groups have been encouraged to develop voluntary flammability requirements through a recognized standards organization. The Upholstered Furniture Action Council (“UFAC”) voluntary industry program of cigarette ignition tests developed in the 1970s is embodied in ASTM E-1353 and other voluntary test methods. Commission staff estimates voluntary UFAC conformance at about 90% of furniture production. The UFAC voluntary program does not address small open flame ignitions. Aspects of the UFAC cigarette ignition resistance test methods, California BHF Technical Bulletins (TB) 116, 117, and 133, and British Standard BS-5852 have been adopted by various consensus voluntary standards organizations and industry groups, including ASTM International, the International Standards Organization, the National Fire Protection Association and the Business and Institutional Furniture Manufacturers of America, and have also been incorporated into some state and local fire codes. Some industry groups have suggested that the Commission should adopt the UFAC program as a proposed rule. As discussed in section G.1 of this preamble, the Commission concludes that mandating the UFAC guidelines would have little effect on reducing deaths and injuries related to upholstered furniture fires.

F. The Proposed Standard

In developing the proposed flammability standard to address ignitions of residential upholstered furniture, the Commission considered the available hazard information, existing standards development research together with the latest CPSC laboratory data, and technical information developed by other organizations. Economic, health and environmental factors were also considered.

1. Scope

The proposed standard contains flammability performance requirements for most residential upholstered furniture. The proposed standard applies to:

• Residential seating products intended for indoor use and constructed with contiguous upholstered seats and backs, such as chairs and sofas (including motion furniture and sleep sofas);

• Some home office furniture sold through retailers or otherwise available for household use; and

• Upholstered furniture used in dormitories or other residential occupancies.

The proposed standard does not apply to:

• Outdoor furniture, such as patio chairs;

• Articles without contiguous upholstered seating surfaces, such as ottomans, decorative pillows or pads, and many office chairs and dining chairs;

• Commercial or industrial furniture not intended or sold for household use;

• Furniture intended or sold solely for use in hotels and other temporary lodging and hospitality establishments;

• Futons, flip chairs, and the mattress components of sleep sofas; and

• Non-furniture juvenile products such as walkers, strollers, high chairs and pillows.

2. General Requirements

The proposed standard addresses resistance to ignition and limited fire growth by means of performance tests for cover fabrics and, alternatively, for

barriers. The principal performance requirements of the proposed standard are intended to reduce the risk of fire from smoldering ignition. If barriers are chosen as the means of compliance, they must meet both small open flame and smoldering resistance requirements. The proposal adapts elements and variations of existing standards, including California Technical Bulletin 117, ASTM E-1353 (tests from the UFAC voluntary industry guidelines) and United Kingdom regulations (based on British Standard BS-5852).

The upholstered furniture tests are conducted using seating mockups of fabric and filling materials. The goal is to reduce the smolder propensity of cover materials and limit the mass loss from combustion (smoldering, melting, or flaming) of the mockup's interior filling materials. Pass/fail criteria are based on maximum acceptable combustion time and mass loss percentages within a 45 minute test period.

3. Cover Fabric Smoldering Resistance Test

In this test, fabrics are tested in combination with a standard polyurethane foam substrate. A lighted cigarette is placed in the seat/back crevice of the mockup and is allowed to burn its entire length. The mockup must not continue to smolder at the end of the 45 minute test or transition to flaming at any time during the test, and the foam substrate must not exceed the mass loss limit of 10%. Ten initial specimens are tested. If the 10 initial specimens meet these criteria, the cover fabric sample passes. If there is a failure in any one of the 10 initial specimens, the test must be repeated on an additional 20 specimens. At least 25 of the 30 specimens must meet the criteria.

4. Interior Fire Barrier Smoldering Resistance Test

In this test, the barrier is placed between a standard foam substrate and a standard cotton velvet cover fabric. A lighted cigarette is placed in the seat/back crevice of the mockup. The foam substrate must not exceed 1% mass loss by the end of the 45 minute test, and the mockup must not transition to open flaming at any time during the test. Ten initial specimens are tested. If all 10 initial specimens meet these criteria, the barrier sample passes. If any one of the ten fails, an additional 20 specimens are tested, and at least 25 of the 30 must meet the criteria.

5. Interior Fire Barrier Open Flame Resistance Test

The proposed standard also contains provisions for the open flame resistance of barriers. In addition to providing protection from small flame ignition, the open flame performance test contributes to the protection of materials from the progression of smoldering to flaming combustion.

In this test, the barrier is placed between a standard rayon cover fabric and standard foam substrate on a metal test frame. An open flame ignition source is applied to the seat/back crevice of the mockup. The mockup must not exceed 20% mass loss by the end of the 45 minute test. Again, 10 initial samples are tested. If there is a failure with any of the 10 specimens, an additional 20 specimens are tested, and at least 25 of the 30 must meet the criteria for the sample barrier to pass.

6. Administrative Requirements

In addition to flammability performance requirements, the proposed standard contains provisions relating to certification and recordkeeping, testing to support guaranties, and labeling of finished articles of upholstered furniture. These requirements are intended to help manufacturers, importers and suppliers ensure that their products comply, and to help the CPSC staff enforce the proposed performance standard. These provisions are contained in Subpart B of the proposed standard.

Under § 8 of the FFA, 15 U.S.C. 1197, producers of finished articles of upholstered furniture, i.e., manufacturers and importers, may rely on guaranties of compliance issued by material suppliers to avoid criminal prosecution in certain instances. However, manufacturers and importers are ultimately responsible under the proposal for compliance of the upholstered furniture products they produce and introduce into commerce. It is unlawful under the FFA to provide a false guaranty. While there are no specific sampling or production testing requirements in the proposed standard, the FFA requires that any guaranties be supported by reasonable and representative tests sufficient to establish that production units of materials meet the applicable tests.

The proposed standard requires that each finished article of upholstered furniture carry a permanent label: (1) Containing a statement certifying that it complies with the standard, identifying the “Type” of furniture (i.e., Type I or Type II); (2) identifying the manufacturer or importer; and (3) specifying the location and month and year of manufacture and model and lot number or other identifier applicable to the item. This information would be required to be separate from other label information. The label would help retailers and consumers identify products in the event of a recall or other corrective action.

G. Response to Comments on the ANPR and Subsequent Submissions

The Commission received 13 written comments during the 60-day formal comment period following publication of the ANPR in October 2003. Since that time, interested parties provided about 20 additional written submissions in the form of letters, position statements or technical presentations at public meetings. Further, the staff held or attended several public meetings with stakeholders to discuss issues of interest.

Many of the public comments addressed similar issues. These issues generally involved: (a) The scope, test methods and acceptance criteria of a possible proposed rule; (b) the potential benefits and costs of various alternatives; and (c) the potential use of flame retardant (FR) chemicals to comply with those alternatives. Some of the comments dealt specifically with the staff's 2001 and 2005 draft standards, options that contained more open flame performance requirements for upholstery materials than the proposed rule. A few of the comments dealt with the staff's 2007 draft proposal, which became the agency's proposed standard. The Commission considered all of the comments received since 2003 in developing the proposed rule.

1. Scope and Test Methods

Comment.

Several industry, government and fire safety organizations provided comments on the general scope of a standard, mainly with respect to cigarette versus open flame ignition performance.

Under the 2003 ANPR, the staff developed multiple draft standards containing both smoldering and open flame requirements. The proposed rule places primary emphasis on smoldering ignition resistance, as a substantial majority of upholstered furniture-related deaths, injuries and property losses result from smoldering fires. Several furniture industry groups commented that the fire risk associated with open flame ignition has become so small that regulation in that area is unnecessary. They also commented that the science of open flame ignition behavior is so complex that substantial further research would be needed to support

any reasonable conclusions about the effectiveness and technical adequacy of any performance requirements. In addition, they opposed open flame ignition requirements on the basis that compliance costs would be unreasonably high. These groups recommended that the Commission proceed with rulemaking on smoldering ignition only, and that CPSC adopt the performance tests in the ASTM/UFAC voluntary guidelines in the proposed rule.

Other stakeholders, including representatives of fire safety organizations, state government and chemical industry groups, recommended that a federal rule contain both smoldering and open flame requirements, and stated that solutions are technically and economically feasible. Some commenters opposed any course that would reduce the current level of safety provided by the existing California regulation, Technical Bulletin (TB) 117. Other industry groups supported adoption of a smoldering standard and eventual consideration of open flame requirements in the future. The California Bureau of Home Furnishings and Thermal Insulation (BHFTI) recommended that CPSC consider adopting elements of the draft revised TB-117 published by BHFTI in 2002.

In 2004, an industry “coalition” of furniture producers and material suppliers developed a set of performance requirements for Commission consideration. The coalition proposal included: a small open flame test for cover fabrics, based on a modification of the Commission's Standard for the Flammability of Clothing Textiles (16 CFR Part 1610); smoldering and open flame tests for filling materials, based on the 2002 draft revision of California TB-117; an open flame test for fibrous (non-foam) “cushion wraps,” based on an existing U.K. regulation provision; ASTM/UFAC smoldering tests for cushion wraps; and an unspecified barrier test to be developed by CPSC. The staff evaluated the industry coalition proposal and questioned the effectiveness of some of the performance elements. Coalition members withdrew support for their proposal in 2005 as the CPSC staff was continuing its evaluation and considering other alternatives.

Response.

The Commission recognizes that estimated residential upholstered furniture fire losses have declined over time, and that relatively few losses—e.g., about 10% of the addressable deaths—are attributable to open flame-ignited fires. Thus, relatively few open flame deaths and injuries could be averted, even under highly effective open flame requirements. The Commission notes, however, that large numbers of deaths and injuries remain. Since a substantial majority of these losses result from cigarette-ignited fires, the Commission agrees that a rule with primary emphasis on smoldering can have substantial safety benefits. Based on CPSC's laboratory research, the Commission also agrees that the ASTM/UFAC test method provides a useful basis for a standard, but does not agree that the ASTM/UFAC tests as implemented in the UFAC voluntary program would adequately achieve those benefits. While UFAC has contributed to fire safety by encouraging the use of smolder-resistant materials, the program allows the use of smolder-prone cover fabrics with polyurethane foam, and allows highly smolder-prone fabrics in combination with more smolder-resistant materials (e.g., polyester batting) underneath. These conforming combinations are not always adequate to prevent fire growth from smoldering ignitions.

CPSC laboratory testing demonstrated that smolder-prone fabrics can defeat the inherent smolder resistance of polyester batting, and that furniture mockup assemblies with highly smolder-prone fabrics can transition from smoldering to flaming combustion over time. Further, some lower-priced furniture may use UFAC-conforming but smolder-prone fabrics without smolder resistant batting. In addition, the UFAC tests may not be adequate to characterize the smoldering behavior of all upholstery materials; for example, UFAC's vertical char length performance metric does not always reflect the downward burning that typically occurs in polyurethane foam fillings. Additionally, the ASTM/UFAC method employs a draft-limiting enclosure that was designed to improve test repeatability but artificially restricts burning of the most smolder-prone fabrics. The non-time-limited UFAC tests may also adversely affect the repeatability of the test results. The Commission concludes that adopting the ASTM/UFAC tests without significant modification would have little effect on currently-produced upholstered furniture, and would therefore probably have negligible safety benefits beyond those already achieved under the voluntary industry program. Thus, the proposed rule has smoldering ignition requirements that are somewhat different from, and more stringent than, those of the UFAC guidelines. The proposed standard also contains open flame performance requirements for barriers; these barriers must protect interior filling materials from smolder-prone fabrics that may otherwise cause furniture to transition from smoldering to flaming combustion.

2. Standardized Test Materials

Comment.

In addition to the CPSC staff's extensive studies on the suitability of various standard test materials, industry groups contributed research and submitted comments on the performance of standard cover fabrics and standard polyurethane foams specified in the CPSC staff's draft standards. Both the staff and industry noted the potential effects of interdependency of standard test materials, and the potential impact on test results of the observed variability in the performance of certain test materials. This variability chiefly related to a standard cotton velvet fabric specified in the open flame tests of the CPSC staff's 2005 draft standard; to a lesser extent, variability was observed in the behavior of the standard FR test foam used in the smoldering tests of the staff's 2005 draft. The comments generally recommended changes to the standard test materials or the test methods to eliminate the undesirable effects of standard material variability.

Response.

The staff's research concluded that the variability identified in the performance of the standard fabric (and, in some cases, the standard non-FR foam) could adversely affect the repeatability and reproducibility of open flame tests, and could yield unacceptably inconsistent results. Similar inconsistencies were observed in the standard FR foam used in smoldering tests. Therefore, the staff revised the qualification requirements for standard test materials to ensure consistency. Further, in view of the hazard data and the complexity (including standard materials variability) of the open flame tests, the proposed rule eliminates the open flame tests for filling materials entirely, and retains standard fabrics for barrier tests only. This approach not only simplifies the proposed standard, but also eliminates the interdependency and variability issues raised by the commenters. The standard cotton velvet test fabric performs consistently in barrier smoldering tests, as does the standard rayon test fabric in barrier open flame tests. Since FR foam would not be needed to comply with the proposed rule, the rule specifies only standard non-FR foam in all tests.

3. Stringency of Requirements

Comment.

Some industry groups opposed the CPSC staff's 2005 draft standard's open flame filling material tests in the absence of an open flame fabric test, and asserted that the 2005 draft's smoldering and open flame filling material requirements were too stringent for some lower-density foams to meet, even with FR treatment. Furniture industry commenters subsequently opposed any requirements that would be more stringent than those of the UFAC guidelines. Many commenters supported the concept of a barrier test option to afford flexibility to manufacturers and fabric suppliers, although some furniture industry groups opposed an open flame requirement for barriers and supported the UFAC smoldering requirement instead. Regarding the staff's 2007 draft proposal that became this proposed standard, some commenters argued that the stringent fabric smoldering requirements would require substantial re-engineering or FR treatment of fabrics. A number of commenters also recommended that CPSC study the effectiveness of reduced ignition propensity (IP), or “fire-safe,” cigarettes before proposing any flammability requirements for upholstered furniture.

Response.

Many of these comments pertained to specific provisions of the open flame requirements of the CPSC staff's 2005 draft standard. The proposed standard does not contain open flame requirements for fabrics or fillings. As noted previously, CPSC's laboratory research on smoldering ignition indicates that several elements of the ASTM/UFAC voluntary approach would not be very effective at reducing the risk. The UFAC guidelines allow smolder-prone combinations of upholstery materials that would not adequately limit fire growth, either from smoldering or transition to flaming combustion. Since the proposed rule relies substantially on cover fabrics or barriers to protect interior filling materials, the proposed standard contains very stringent smoldering requirements, and requires that barriers provide protection regardless of cover fabric ignitability.

The Commission agrees that a significant proportion of predominantly cellulosic fabrics (i.e., chiefly cotton fiber content) would have to be modified or eliminated under the proposed standard. The Commission notes that these fabrics are the most smolder-prone materials used in upholstered furniture, and that many smolder-prone fabrics can sometimes overwhelm the inherent smolder resistance of synthetic filling materials like polyurethane foam or polyester batting. Thus, the proposed requirements are applied to those materials whose ignition behavior is the primary contributor to the risk.

The proposed standard would not prohibit fabric suppliers from using FR-treated fabrics to comply. However, furniture and textile industry representatives have stated a desire to avoid such products for aesthetic and cost reasons. Given the availability of non-FR alternatives, it is unlikely that fabric suppliers would use the FR treatments they said consumers would reject.

The Commission agrees that reduced ignition propensity cigarettes may be an effective means of reducing consumer product-related smoldering fires. Such reductions would likely occur irrespective of CPSC action on upholstered furniture. An increasing number of states (and Canada) have “fire safe cigarette” laws that now require or will require that only reduced-IP cigarettes be available for sale. Complying cigarettes would likely reduce, but would not eliminate, the risk of smoldering ignited upholstered furniture fires. The extent of any such reduction is unknown. The staff has initiated a study to review available state data and to conduct laboratory tests to evaluate the reduction in smoldering ignition propensity associated with reduced-IP cigarettes compared to conventional cigarettes. This work will help the Commission evaluate the potential effect of reduced-IP cigarettes on upholstered furniture fire losses.

4. Large Scale Validation Testing

Comment.

Some stakeholders recommended that CPSC establish a correlation between its bench scale tests in the proposed rule and the performance of complying materials in larger or “full” scale tests that more reasonably represent the seating areas of finished articles of upholstered furniture. These large scale tests would help validate the results and potential effectiveness of the bench scale tests.

Response.

The Commission agrees that large scale testing is a valuable source of information to help demonstrate the increased safety the proposed standard would provide. To supplement the CPSC staff's bench scale testing and limited large scale testing performed previously, the staff plans to sponsor such large scale tests. The Commission can use the results of these tests in developing a possible final rule.

5. Potential Benefits and Costs

Comment.

Some industry groups submitted comments about the CPSC staff's draft preliminary regulatory analysis of potential benefits and costs associated with various regulatory alternatives. Most of these comments were from organizations that opposed various aspects of the CPSC staff's 2005 draft standard; some of the comments related to the staff's draft proposal that became the proposed standard.

The comments on the staff's analysis of the 2005 draft standard generally asserted that the staff had overestimated potential benefits and understated potential costs. A 2006 furniture industry report on the staff's analysis of the 2005 draft standard and alternatives criticized the statistical methodology used to develop national fire loss estimates, and recommended different methods that would generally result in lower estimates of potential benefits of a flammability rule. The report also questioned other aspects of the staff's estimation of potential economic benefits of a standard, positing that staff overstated benefits by using effectiveness estimates and value-of-life estimates that were too high, discount rates that were too low, and incorrect assumptions about the distribution of smolder-prone furniture fabrics among smoking vs. non-smoking households.

The 2006 industry report also asserted that the staff understated costs to filling material suppliers and furniture manufacturers and importers, and recommended that the staff's sensitivity analysis consider all combinations of factors affecting benefits and costs unless those factors were mutually exclusive. Manufacturers of polyurethane foam raised some of the same cost issues, and discussed anticipated difficulties in producing consistently-complying foams at the lower densities often used in upholstered arms and other areas of furniture.

Regarding the CPSC staff's 2007 draft proposal, some textile industry representatives criticized the emphasis on cover fabric performance, and expressed concern that the standard would not regulate filling material performance. They also expressed concern that difficulties in modifying many fabrics, combined with the cost of “double-upholstering” furniture to incorporate interior barriers, may lead suppliers to use FR treatments to comply. One report prepared for an environmental group recommended that CPSC include in its analysis of the 2007 draft estimates of economic losses from

increased cancer risks associated with FR filling material additives.

Several stakeholders recommended that CPSC consider the effect of reduced ignition propensity (IP), or “fire-safe” cigarettes on the potential benefits of a possible upholstered furniture flammability standard. One report prepared for an environmental group presented an alternative calculation of benefits incorporating some different assumptions about reduced-IP cigarette effectiveness than those made by the CPSC staff in 2006. Some industry commenters suggested that as reduced-IP cigarettes came into wider use, a standard for upholstered furniture would no longer have net benefits to the public.

Response.

Regarding fire loss estimation methodologies, the CPSC staff noted several biases and errors in the industry report's approach that would misrepresent the estimates of fire losses. The 2006 industry report's criticism of the staff's method did not consider the proper allocation of fire incidents with unknown fire causes. Further, the indirect estimating method recommended by the industry report incorrectly used estimates of the number of fires to estimate death and injuries, thereby introducing bias and understating deaths. The CPSC staff's method correctly used death and injury counts weighted with probability-based estimates for fire deaths and injuries. Another method suggested by the industry report wrongly excluded some in-scope deaths from the body of data used to make the estimates. The use of these recommended alternative methods would significantly understate fire losses, and would thereby understate the potential benefits of a flammability rule.

Regarding benefits projections, the preliminary regulatory analysis of the proposed rule estimated the monetary value of potential benefits using estimates of effectiveness based on CPSC laboratory tests of upholstered furniture mockup assemblies constructed with ignition resistant fabrics or barriers, and using adjustments to reflect the projected mix of products on the market and other factors. Large scale tests will help support the effectiveness estimates. However, the Commission staff has ample experience to date with upholstery material testing to estimate that the proposed rule would likely be highly effective (about 60%) at reducing fire deaths, injuries and property damage. Even considering the effectiveness estimates for the CPSC staff's 2005 draft standard, there is no basis for applying effectiveness rates for the U.K. regulations to a CPSC rule. Further, the sensitivity analysis in the preliminary regulatory analysis accounts for uncertainty in the estimates.

The Commission staff estimated the present value of future safety benefits using discount rates (3% and 7%) recommended by the Office of Management and Budget in its guidance on regulatory analyses. Also, CPSC's statistical value of life estimate ($5 million) and sensitivity analysis range ($3-7 million) is consistent with values cited in the economic literature and widely used in regulatory decision-making.

Regarding the distribution of upholstered furniture constructed with smolder prone fabrics among smoking vs. non-smoking households, the preliminary regulatory analysis assumed that furniture fabric types are distributed evenly among households. Smolder prone fabrics are often, but not always, used on the very high-priced, decorator furniture more commonly found in higher-income households that tend less often to be smoking households. However, anticipated market trends include potential future increases in predominantly-cotton fabrics in more moderately-priced furniture, especially among imports, which tends to be lower in price than domestic products. To the extent that furniture with smolder prone fabrics is more often found in higher-income households with lower smoking prevalence, the benefits of a flammability rule could be reduced somewhat. The preliminary regulatory analysis notes in its sensitivity analysis that the likely impact on benefits would be small.

The sensitivity analysis in the preliminary regulatory analysis considers the impact of a variety of factors on potential benefits and costs. Varying more than one factor at a time is generally appropriate when those factors are highly correlated, rather than whenever they are not mutually exclusive, as the 2006 industry report suggested. The sensitivity analysis does take into account some combinations of factors, but not all factors that could conceivably affect benefits and costs. However, even if all of the combinations of possible factors were considered together, estimated net benefits of the proposed standard would still total $100 million or more from a year's production of complying upholstered furniture.

The staff considered likely cost impacts on fabric, filling material and other upholstery material suppliers in analyzing the potential impacts of the proposed standard. Cost estimates were generally reported directly as provided by firms in the industry sectors affected although some cost estimates varied significantly among firms. The preliminary regulatory analysis recognized several areas of cost concern, including low-density polyurethane foam and loose filling materials (for the staff's 2005 draft standard) and certain 100% cotton fabrics (for the 2007 draft). The staff analysis noted that while most upholstered furniture fabrics would meet the proposed standard without modification, more than half of all predominantly cellulosic fabrics may fail the proposed standard fabric smoldering test. These smolder-prone fabrics are typically used with synthetic filling materials that would otherwise be generally smolder resistant; thus, the proposed standard targets those fabrics contributing most to the risk of smoldering ignition.

The staff also noted that some of the more expensive decorator fabrics that would fail the proposed fabric smoldering test already are used in furniture that employs multiple layers of upholstery materials, or “double upholstering.” Decorative fabric suppliers have long supported a barrier option for use with non-complying fabrics. For most articles of upholstered furniture, the barrier option incorporated into the proposed standard would involve substituting complying barriers for existing interior fabrics or battings; this would amount to a “drop-in replacement” of existing components for most barriered furniture, and would not require significant additional assembly labor costs.

The preliminary regulatory analysis estimates costs based on the assumption that some or all non-complying fabrics not used with barriers would be FR treated; however, it is unlikely that a significant proportion of fabrics would actually be treated; thus, material costs may be lower than estimated in the analysis. Compliance costs associated with re-engineering some heavier-weight, 100% cellulosic fiber fabrics may be significant for some firms, although fiber content modifications are made routinely by producers (sometimes as often as every six months) to reflect style trends in the market. Blended-fiber fabrics in particular could probably be readily modified without difficulty or significant disruption.

Under the staff's draft 2005 standard, FR foam fillings would likely be used to comply. One of the FRs currently used in foams meeting the existing California TB-117 may pose cancer and non-cancer chronic health risks. Pending

further study of these and other FR chemicals, the preliminary regulatory analysis of alternatives assumed that hazardous FRs would not be used to comply, and therefore did not include a calculation of possible disbenefits associated with potential use of any potentially hazardous filling material FRs. The proposed standard would not require the use of any FRs in foam or other interior filling materials.

The Commission considered the potential impact of reduced-IP cigarettes, and continues to study this matter. State requirements for such cigarettes may reduce upholstered furniture fire losses over time irrespective of CPSC action. The extent of the reduction is unknown. The preliminary regulatory analysis does specifically account for possible risk reductions associated with reduced-IP cigarettes. If, for example, reduced-IP cigarettes reduced the level of benefits of the proposed rule to half the estimated level, then projected net benefits would be reduced from $367-387 million to $155-177 million per year's worth of complying furniture production. Even at a 70% benefit reduction, estimated net benefits of the proposed rule would still approach $100 million.

6. Potential Use of FR Chemicals

Comment.

The Commission received a number of comments either opposing or supporting the potential use of FR chemical technologies to meet a possible flammability rule. Most of these comments related to the staff's previous, 2005 draft standard, which would have required that resilient, fibrous and loose filling materials (typically made of polyurethane foam or polyester fiber) be open flame resistant. Some comments specifically opposed the use of polybrominated diphenyl ethers (PBDEs), and cited studies on the potential health and environmental risks related to these compounds. At least one of the major filling material FRs, penta-BDE, that was previously used to meet California TB-117's open flame requirements, has been discontinued. While most fillings would be FR-treated under the 2005 draft, the proposed standard does not contain filling material requirements, and FR additives would not be needed to comply.

Some environmental groups opposed any new regulations that may add to the environmental burden of FR chemicals, especially halogenated FRs containing bromine or chlorine. They contended that since some FRs are persistent in the environment, bioaccumulative in animals and potentially toxic to humans, and since there is a lack of data on some aspects of the potential effects on human health and environmental risks, the Commission should not encourage the use of these chemicals. Some of these groups supported the furniture industry position that CPSC should impose only smoldering ignition requirements, on the presumption that FRs would not be needed to meet these requirements. The environmental groups strongly supported the staff's 2007 draft proposal that became this proposed standard.

Furniture and filling material producers opposed significant increases in FR usage on the basis that their workers could be exposed to more FRs released from component materials. They were also concerned that state and local environmental regulations may curtail the availability of economically feasible FRs and may adversely affect manufacturers' ability to recycle scrap materials. Furniture and fabric manufacturers also contended that, in view of recent adverse publicity, consumers would prefer not to risk exposure to potentially toxic FRs. Some representatives of fabric suppliers have also expressed concern that any smolder resistance requirements more stringent than those in the UFAC voluntary guidelines would force many firms to use FR treatments on predominantly cotton fabrics to comply.

Chemical producers stated that safe and effective FR solutions are available to address the furniture risk. They noted that non-halogenated alternatives for filling materials are currently being offered or developed, as are “inherently-FR” fiber barriers that do not present a significant likelihood of consumer exposure.

Response.

CPSC developed the proposed standard mindful of the continuing uncertainty about potential health and environmental effects of FR chemical usage, with an objective of achieving significant reductions in fire deaths and injuries from upholstered furniture fires caused by smoking materials while minimizing reliance on FR additives in fabrics and filling materials to meet that objective. While the available scientific data are sufficient to show that some FRs would not present significant health or environmental risks, the Commission agrees that insufficient data are available to be reasonably sure that other FRs would not present health risks if used in upholstered furniture. The staff's health risk assessment for foam filling materials concluded that the polyurethane foam FR most widely used to meet California TB-117 may not present chemical risks to consumers but identified significant data gaps; the risk assessment further indicated that another currently used filling material FR may present both cancer and non-cancer risks to consumers. On the other hand, the CPSC staff's health risk assessment for barriers concluded that several commercially available technologies, including inherently-FR fiber products, could be used without presenting appreciable health risks to the public.

Under the proposed standard, neither fabrics nor filling materials would need to incorporate FR additives to achieve compliance. While FR-treated fabrics would not be prohibited, many fabric suppliers have indicated they would likely either modify the fiber content or construction of their most smolder prone fabrics, or continue to offer non-complying fabrics for use exclusively with complying barriers in the finished article of furniture. Thus, the Commission anticipates that FR fabrics would be the least likely means of compliance with the proposed rule. Barriers could incorporate FR treatments, but barrier suppliers have reported that they would likely offer inherently-FR fiber materials that do not pose a risk of potential exposure for upholstered furniture applications, similar to those products designed to meet the Commission's open flame rule for mattresses (16 CFR part 1633). Barriers are projected to be used in only about 5% of all upholstered furniture; most of this usage would be in designer or higher-priced furniture for which the relatively higher cost of barriers would not be a significant factor.

The Commission plans to monitor the progress of ongoing studies on FR chemicals and to consider the results of those studies as the regulatory process continues. At the request of the staff, the National Toxicology Program (NTP) of the Department of Health and Human Services has undertaken a review of several FRs that could be used to meet CPSC flammability rules. The NTP review will be a relatively long-term project that contributes to the overall level of knowledge about FR chemicals among scientists and regulators.

H. Preliminary Regulatory Analysis

The Commission has preliminarily determined to issue a rule establishing a flammability standard addressing the ignition of upholstered furniture. Section 4(i) of the FFA requires that the Commission prepare a preliminary regulatory analysis for this action and that it be published with the proposed rule. 15 U.S.C. 1193(i). The following discussion, extracted from the staff's memorandum titled “Preliminary

Regulatory Analysis of a Draft Proposed Flammability Rule to Address Ignitions of Upholstered Furniture,” addresses this requirement.

1. Introduction

The history of this rulemaking is discussed in Section A, Background, of this preamble. This Preliminary Regulatory Analysis discusses the impacts of provisions specified in the Commission's proposed standard for upholstered furniture. It provides information on the products and industries that are likely to be affected by actions taken to reduce upholstered furniture fires. The analysis also discusses potential costs and benefits associated with requirements of the proposed standard and reasonable alternatives. This analysis also discusses potential effects on small firms and other market impacts.

2. The Proposed Standard: Scope and Provisions

The proposed standard contains smoldering ignition performance requirements for cover fabrics, and smoldering and open flame performance requirements for interior fire barriers (if they are used as the method of compliance). The proposed standard applies to finished or ready-to-assemble articles of upholstered furniture (such as upholstered sofas, loveseats, sofa beds, rockers, recliners, and other chairs) that are: primarily intended for indoor use in residences; constructed with an upholstered seating area, comprised of a contiguous upholstered seat and back or arm(s); and manufactured or imported after the effective date.

The proposed standard offers manufacturers two alternative methods to produce complying furniture. Furniture items can comply by being made with upholstery cover materials that pass the cover material smoldering ignition resistance test (designated as “Type I upholstered furniture” in the proposed standard). Alternatively, manufacturers may comply with the proposed standard by using a barrier material under the upholstery fabric that passes the standard's applicable barrier tests (“Type II upholstered furniture”). This option allows manufacturers to use non-complying upholstery fabrics.

3. Products and Industries Potentially Affected

The largest class of furniture products that would be affected is upholstered furniture on wood frames and dual purpose sleep furniture such as sofa beds, commonly bought for use in living rooms and family rooms. Other types of affected products include upholstered metal frame, reed, and rattan furniture.

Products referred to as “Household Upholstered Furniture” by the Census Bureau are classified in code 337121 of the North American Industrial Classification System (NAICS). This classification includes production of upholstered furniture on frames made of wood, metal, or other materials, as well as dual-purpose sleep furniture, such as convertible sofa beds. The 2002 Economic Census reports that 1,686 U.S. companies (with 1,946 establishments) manufactured upholstered household furniture or dual-purpose sleep furniture as their primary product.

Many other firms may also produce upholstered furniture as secondary products.

U.S Census Bureau, U.S. Department of Commerce, 2002 Economic Census, report EC02-311-337121, “Upholstered Household Furniture Manufacturing: 2002,” September 2004.

The Economic Census reports that the value of shipments of upholstered household furniture by U.S. firms in 2002 was $10.3 billion. The Annual Survey of Manufactures reported value of product shipments of $10.0 billion in 2003 and $9.55 billion in 2004.

The value of product shipments for 2005 was reported by the Census Bureau to have totaled $9.9 billion.

U.S Census Bureau, U.S. Department of Commerce, Value of Product Shipments: 2005, Annual Survey of Manufactures, November 2006.

Although there are a large number of upholstered furniture manufacturers, the top four companies accounted for nearly 35 percent of the total value of household upholstered furniture shipments in 2002 (the latest year for which industry concentration ratio data are available); the 50 largest companies accounted for about 67 percent.

Reports from the trade press indicate that the industry has become more concentrated in the last ten years. Several firms have ceased operations; others have merged with larger companies through buyouts. The consolidation included Furniture Brands International's acquisition of HDM Furniture Industries (which included Henredon and Drexel Heritage) in 2001, and La-Z-Boy's acquisition of Ladd in January 2000 and Bauhaus and Alexvale in 1999. La-Z-Boy is the number one upholstered furniture manufacturer (by dollar volume), and Ladd, Bauhaus, and Alexvale all previously ranked in the top 30. Furniture Brands International is the second-leading domestic manufacturer of upholstered furniture, and companies it acquired were previously part of number four-ranked LifeStyle Furnishings, International, Ltd.

U.S Census Bureau, U.S. Department of Commerce, 2002 Economic Census, report EC02-31SR-1, “Concentration Ratios: 2002,” May 2006.

The industry also includes many small companies and establishments. The 2002 Economic Census reports that only 29 percent of upholstered furniture establishments (564 of 1,946) had 20 or more employees, and only 10 percent (200 establishments) had 100 or more. By some measures, such as the U.S. Small Business Administration's (SBA's) definition for qualification for small business loans, a furniture manufacturing company is considered to be “small” if it has fewer than 500 employees (at all of its establishments). This definition encompassed more than 97 percent of firms in the industry in 2002.

Based on 2002 firm size data compiled by the United States Small Business Administration's Office of Advocacy which is available online at

http://www.sba.gov/advo/research/data.html

Exports of upholstered furniture had a value of about $285 million in 2005, or almost 3 percent of the total value of shipments.

The value of imports of products categorized by the Census Bureau as NAICS 337121 was $2,792 million in 2005.

Therefore, there were net imports of about $2.5 billion. With estimated domestic shipments of $9.9 billion, these net imports resulted in total apparent consumption of upholstered furniture in 2005 (domestic shipments plus imports, minus the value of exports) of about $12.4 billion.

U.S. Department of Commerce data.

U.S. Department of Commerce and U.S. International Trade Commission data (c.i.f. cost basis).

Imports have grown in recent years, accounting for about 22 percent of the value of total apparent consumption of residential upholstered furniture in 2005. By way of comparison, about 10 percent of the value of apparent consumption of upholstered household furniture in 1999 was imported. The leading country of origin is China, which accounted for about 52 percent of the value of imports in 2005 and nearly 63 percent of the value of imports in 2006. Mexico accounted for about 11 percent of imports in 2006; Italy about 8 percent, and; Canada about 5 percent. These four countries accounted for 86 percent of the total value of imported upholstered furniture in 2006.

The importance of China as a source for imports has grown significantly in recent years. China supplanted Italy as the leading country of origin in 2003, and by 2006 the value of imports from China was almost 6 times that of the second-ranked country of origin, Mexico. Italy had been the number one source for upholstered furniture imports

for many years. The majority of units from both China and Italy in 2004 reportedly were upholstered in leather.

Although much of the gain in China's market share has been at the expense of Italian imports, some of the furniture imported from China is from plants that have been established by several major Italian firms. China has been the leading source of wood (non-upholstered) furniture imports and its growth as a source of upholstered furniture is expected to continue.

Industry analyst, Jerry Epperson, reported in Furniture Today, December 12, 2005. p. 66.

In addition to affecting manufacturers of residential upholstered furniture typically found in living room and family rooms, the proposed standard also includes dining room and kitchen chairs within its scope if they are made with contiguously upholstered seats and backs. Similarly upholstered desk chairs purchased for household use are also covered by the standard. Dining chairs are generally products of firms classified in the wood household furniture industry, NAICS 337122. The Economic Census reports that 4.8 million wood dining room chairs were shipped in 1997, with a value of shipments totaling about $526 million. In 2002, shipments fell to 2.9 million chairs, with a value of about $446 million. The decline in domestic shipments is attributable to significant increases in imports of wood furniture from China and other countries.

Census data are not reported separately for upholstered and non-upholstered dining chairs. In 1994, an industry-sponsored study surveyed participants in the voluntary industry program to improve the cigarette ignition resistance of furniture that was developed by the Upholstered Furniture Action Council (UFAC). Among the firms surveyed were manufacturers of upholstered dining room and kitchen seating. The study report estimated that the total value of shipments of such furniture that complied with the UFAC Program (and, therefore, had upholstered seats) was about $250 million for 1993.

Based on the value of 1992 shipments ($580 million), perhaps 3 to 4 million upholstered dining chairs were shipped by these UFAC participants. A great majority of these items may not have had upholstered backs, or they had upholstered backs that were not contiguous with upholstered seats. Other firms that are not participants in the UFAC Program also manufacture upholstered dining furniture. Given the limitations of the market data, the number of dining chairs produced annually that fall within the scope of the proposed standard cannot be estimated with much precision, although the total number of units is thought to be relatively small.

Heiden Associates, Inc., “Report on Survey of UFAC Members re: Compliance with Upholstered Furniture Cigarette Ignition Flammability Standard,” December 15, 1994.

Annual domestic retail sales of all types of living room and family room upholstered furniture total about 30 to 33 million units with a value of over $20 billion. Furniture manufacturers, especially smaller firms, commonly market their products through independent sales representatives who provide information on the market, and get and service new retail accounts for manufacturers. Recently, some manufacturers have reduced their reliance on independent representatives by employing their own salespeople.

Besides purchasing from manufacturers through independent sales representatives or the manufacturers' own sales staff, retailers may purchase furniture from wholesale furniture distributors. These wholesalers purchase from perhaps 25 to 30 manufacturers of different types and styles of furniture. The sales staffs of the wholesalers then call on retailers within their areas. Dealing through local wholesalers that stock an assortment of furniture, and that also offer competitive prices, credit, and other services, is advantageous to many retailers, particularly smaller firms.

Handbook of Furniture Manufacturing & Marketing, Volume 9, Wholesaling, AKTRIN Research Institute and High Point University, May 1994.

According to the 2002 Census of Retail Trade, 19,403 retail establishments carried upholstered furniture as a product line.

Retail prices of upholstered furniture fall into a very broad range, depending on materials and manufacturing techniques used. Larger retailers are more likely to purchase directly from furniture manufacturers, and smaller firms are more likely to purchase through wholesale distributors. Increasingly in recent years, retailers have reportedly devoted more floor space to private labeled furniture imported directly from foreign manufacturers. In response, several of the larger domestic furniture manufacturers have opened or expanded their own retail outlets.

U.S Census Bureau, U.S. Department of Commerce, 2002 Economic Census, report EC02-441-09 “Furniture Stores: 2002,” August 2004.

A review of trade publications indicates that approximately 100 to 200 domestic manufacturers derive a significant share of their revenues from fabric for residential upholstered furniture.

This number includes textile mills that produce finished upholstery fabric and textile finishers that purchase unfinished goods and perform additional processes, such as printing and dyeing. Like the upholstered furniture manufacturing industry, the 1990s saw consolidation of firms specializing in upholstery fabric production, with larger firms buying out competitors or divisions of competitors. However, in just the last few years the U.S. industry has been shaken by the decreased demand for domestically-produced fabric as a result of increased competition from imported upholstery fabric, the increased popularity of leather upholstery, and the dramatic increase in consumption of upholstered furniture imported from China. One of the largest marketers of upholstery fabrics in the U.S. reported that the trend to greater foreign competition and the entry of more converters of upholstery fabric (companies that purchase and resell fabrics) has resulted in greater fragmentation of the upholstery fabric industry in recent years, with lower barriers to entry, and an increase in competition based on price.

Including the Directory of Manufacturers published by the former industry association, the American Textile Manufacturers Institute (ATMI).

Culp, Inc., Annual Company report for the fiscal year ended April 29, 2007.

Interior fabric revenues of the top 10 firms totaled more than $1.9 billion in 2002, based on a trade press survey.

These revenues included sales of fabrics other than those used in residential upholstery. A similar survey found that the top 10 upholstery fabric mills had combined revenues from interior fabric shipments of $2.4 billion.

In addition to declining sales for the leading U.S. upholstery fabric manufacturers, the difficult state of the industry is evidenced by recent bankruptcies of firms that were once industry leaders, such as Joan Fabrics (previously the number one upholstery manufacturer) and Quaker Fabric (previously the number three firm). Both of these firms ceased operations and their production facilities were liquidated in 2007.

“U.S. fabric producers still standing despite import wave.” Furniture/Today, Cahners Publishing, Greensboro, NC, June 2, 2003.

“Mastercraft buy puts Joan at top.” Furniture/Today, Cahners Publishing, Greensboro, NC, June 1998.

Textile mills that make upholstery fabrics as their primary products are included in the North American NAICS code 313210. Of 663 firms in NAICS 313210 in 2002, only 63 (about 10 percent) had 500 or more employees. About 65 percent of the firms had fewer

than 20 employees.

The SBA considers firms with fewer than 1,000 employees to be small businesses for the purposes of programs administered by that agency. Although these data are indicative of the sizes of firms involved in the production of furniture upholstery fabrics, NAICS 313210 encompasses many firms that produce fabrics other than furniture upholstery. Nevertheless, it is likely that nearly all manufacturers of upholstery fabrics could be considered small businesses under SBA guidelines.

Based on 2002 firm size data compiled by the United States Small Business Administration's Office of Advocacy which is available online at

http://www.sba.gov/advo/research/data.html

Fabric finishers also tend to be small. Finishers are firms that receive unfinished fabrics (“greige goods” or “gray goods”) and perform additional manufacturing processes (e.g., printing, dyeing, backcoating, needle-punching, and stain-guarding). Fabrics may be purchased by the finishers, or finished under contract to other firms that supply the fabrics. Fabric finishers are classified in NAICS code 313311. Of 1,016 broadwoven fabric finishing firms in NAICS 313311 in 2002, only 30 (3 percent) had 500 or more employees.

Only a few firms currently apply FR treatments to upholstery fabrics.

Ibid.

The U.S. Census Bureau reported that U.S. upholstery fabric production in 2004 was 284 million square yards (which is the equivalent of 189 million linear yards).

This production was 43 percent lower than 2002's reported production of 499 million square yards (332 million linear yards) of upholstery fabric.

The number of looms in operation for the production of these fabrics totaled 2,610 at the end of 2004, down 20 percent from 3,098 looms at the end of 2002. The major end-use markets for upholstery production are in upholstered furniture and automobile manufacturing. Upholstery fabrics are also used in the manufacture of window treatments and other home textiles. Based on a survey of upholstered furniture manufacturers by Ciprus, Ltd., about 233 million linear yards of upholstery fabric were consumed in the production of household furniture in 2001.

This total does not include leather and vinyl upholstery, which are estimated to have comprised about 30 percent of all furniture upholstery materials used in 2001. Therefore, total upholstery use for the domestic manufacture of residential upholstered furniture was about 333 million linear yards. Estimates of total annual upholstery fabric consumption based on average requirements for chairs and sofas/loveseats are 225 million linear yards.

U.S. Census Bureau.

Current Industrial Reports, Broadwoven Fabrics (Gray):

2004. MQ313T(04)-5. June 2005.

U.S. Census Bureau.

Current Industrial Reports, Broadwoven Fabrics (Gray):

2002. MQ313T(02)-5. June 2003.

Ciprus Limited, LLC.

The North American Market for Contract & Residential Upholstery Fabric,

2001.

According to industry sources, an average of approximately 7 linear yards of fabric is needed to upholster chairs and 11 to 15 yards are needed for sofas. Based on about 31.5 million annual unit shipments (of which perhaps about 53 percent are sofas, sofabeds, and loveseats and about 47 percent are other chairs), estimated annual upholstery material requirements are about 321 million linear yards (about 217 million yards for sofas, sofabeds and loveseats plus 104 million yards for chairs).

The U.S. Census Bureau's Economic Census report, Upholstered Household Furniture Manufacturing: 2002, included information on the costs of upholstery fabrics and other materials used in the production of upholstered household furniture in that year. The report placed the delivered cost of woven cotton upholstery fabrics (excluding ticking) at $312 million and the delivered cost of other woven upholstery fabrics, such as those made of rayon, nylon, and polyester (excluding ticking) at $802 million.

The combined total delivered cost of upholstery fabric of $1,114 million was about 22 percent of the total delivered cost of all materials used in upholstered furniture manufacturing in 2002 (which was, according to the Census Bureau, $5,107 million). Other upholstery cover materials include leather, which is not reported as a separate material category by the Bureau of the Census, and coated and laminated fabrics, which had a delivered cost of about $185 million in 2002. In its 2007 Annual Report, La-Z-Boy, the largest manufacturer of upholstered furniture in the U.S., reported that purchased cover materials (primarily fabric and leather) accounted for about 28 percent of the total cost of raw materials for its upholstery group.

U.S. Census Bureau,

2002 Economic Census, Upholstered Household Furniture Manufacturing:

2002, EC02-311-313311. September 2004.

La-Z-Boy, Inc.

Annual Report for the Fiscal Year Ended April 28, 2007

(Form 10-K.) Page 5.

Until recent years, relatively little upholstery fabric was imported. A report by Keyser Ciprus, Ltd., estimated that 8 million linear yards of residential upholstery fabric were imported in 1997. That accounted for approximately 2 percent of total consumption of upholstery fabric for residential furniture production in that year.

However, as noted above, foreign upholstery fabric production facilities (located primarily in China) have expanded operations and imports of upholstery fabrics have grown substantially.

Keyser Ciprus Limited, op. cit., p. 40.

Much of the foreign production is from facilities that are owned or operated in partnership with U.S. textile firms. For example, Culp, Inc., reported that almost 60 percent of their sales of upholstery fabrics in their fiscal year ended April 29, 2007, consisted of fabrics produced in plants outside the U.S., compared to 17 percent of sales just two years before.

Culp owns and operates four upholstery plants in Shanghai, China, and markets other fabrics from third party sources which are also located in China. The firm only has one remaining upholstery fabric plant in the U.S., down from fourteen in 2000.

Culp's experience in shifting production to foreign plants has also been reported by other U.S. upholstery fabric manufacturers. In January 2007 Richloom Fabrics Group shifted production of its Berkshire Weaving upholstery line from its South Carolina plant to a facility in Shanghai.

Quaker Fabric Corporation also entered into business agreements in recent years with Asian firms to produce fabrics it designs. Quaker estimated that, industry-wide, about 42 percent of total domestic upholstery fabric sales (excluding automotive fabrics) were imported in 2004, versus only 11 percent in 2002. The company's management believed it was likely that the trend continued, and it estimated that about 60 percent of furniture upholstery fabric sales were imported by the end of 2006.

As noted above, Quaker Fabric, which had long been a major U.S. producer of upholstery fabric, could not successfully adjust its operations to meet the recent market shifts, and the firm liquidated its operations in 2007.

Culp, Inc. Annual company report for the fiscal year ended April 29, 2007. (Reportedly includes fabrics produced at Culp's Shanghai manufacturing plant and production sourced from other Asian firms.)

Culp, Inc. Annual company report for the fiscal year ended April 23, 2000.

Andrews, Susan M. “Richloom moves production to China.” Furniture/Today, December 18, 2006.

Quaker Fabric Corp.

Annual Report for the Fiscal Year Ended December 30, 2006

(Form 10-K.).

At least until recent years, exports of upholstery fabric were significant for many U.S. manufacturers. In the late 1990s as much as 20 percent of the upholstery fabric production by U.S. manufacturers in recent years may have been exported. As noted above, more upholstery fabric is being imported from China and other foreign sources in more recent years, and some major U.S. fabric

manufacturers have established production facilities in China, or have established business relationships with Chinese firms to produce fabrics to their specifications and designs. These market changes could be expected to reduce exports by domestic firms from previous levels.

There is a growing practice, especially for leather, to purchase fully cut and sewn parts from areas outside of the United States including but not limited to: Argentina, Brazil, China, Italy, Thailand and Uruguay. This trend should continue given the lower labor costs in some of these areas and other existing economic conditions. La-Z-Boy reports that importing cut and sewn leather parts results in savings of 10 to 20 percent compared to domestic purchases and fabrication of these parts.

Cut and sewn “kits” reportedly are manufactured to the specifications of furniture manufacturers at facilities maintained by foreign fabric producers. Culp reports that it rapidly expanded its cut and sew operations in its Shanghai plants.

La-Z-Boy.

op. cit.,

p. 4.

Culp, Inc. Annual Company report for the fiscal year ended April 29, 2007.

CPSC-sponsored surveys of furniture manufacturers in 1981, 1984, and 1995, and commercial surveys in 1997, 2001, and 2006

provided information on two characteristics of fabrics: fabric type and principal fiber (or material) type. Fabric Type refers to commonly-accepted descriptions of the ways in which fabrics are manufactured or of their distinctive characteristics. For the period covered by these surveys, manufacturers increased their use of jacquard and dobby fabrics, and decreased their use of velvet fabrics.

Usage of cotton prints and flocks fluctuated within fairly narrow ranges during the period, according to the surveys.

Keyser-Ciprus, Ltd. survey (1997) and Ciprus Limited, LLC, surveys (2001 and 2006).

“Jacquards” and “dobbies” refer to the types of looms and weaves used to produce fabrics. Brocades, damasks, velvets, tapestry weaves, and matelasses are often jacquard-woven. Dobbie looms enable weaving of small, geometric figures as a regular pattern. Dobby looms produce patterns that are beyond the range of simple looms, but are somewhat limited compared to a jacquard loom, which has a wider range of pattern capabilities.

Fiber (or material) Type refers to the fibers or materials used in the manufacture of the fabrics or upholstery. Most upholstery fabric fibers are classified as cellulosic (e.g. cotton and rayon) or thermoplastic (e.g., polyester, polyolefin, and nylon); other materials used to make upholstery include vinyl (which is coated on a base fabric), wool, and leather. Based on the 2006 Ciprus Limited survey, cellulosic fabrics currently account for about 25 percent of upholstered furniture upholstery covering materials. Thermoplastic fabrics account for 45 percent; leather, wool and vinyl-coated fabrics account for about 30 percent (mostly leather).

Review of the data on material types from the surveys conducted since 1981 indicates that the most notable changes over the years have been the increase in use of leather at the expense of both cellulosic and thermoplastic fibers. The Ciprus survey in 2001 found that about 30 percent of furniture covering materials used in that year was leather, significantly greater than found in the earlier surveys.

Fabrics made from predominantly cellulosic fibers include heavier-weight fabrics (such as cellulosic jacquards and velvets) and lighter-weight fabrics (mainly cotton prints). Analysis of survey data since 1981 indicates that heavier cellulosic fabrics have usually comprised about 15 to 20 percent of all upholstery covering yardage.

Ciprus Limited.

op. cit.

4. Characteristics of Furniture in U.S. Households

The number of furniture units in use is estimated with the CPSC Product Population Model, based on available annual sales data and industry estimates of the average product life of furniture.

Estimates are for sofas, loveseats, armchairs, recliners, convertible sofas and other upholstered furniture commonly found in residential living rooms, family rooms, and guest rooms.

M.L. Lahr and B.B. Gordon, Final Report on Product Life Model Feasibility and Development Study, Battelle Columbus Laboratories, July 14, 1980.

Sales are defined as shipments from U.S. manufacturers plus net imports. Annual shipment data are available from the Economic Census published every five years (i.e., 2002, 1997, 1992 * * *) by the Bureau of the Census. For upholstered wood furniture and dual-purpose sleep furniture, the Economic Census usually provides information on unit shipments, by type (such as sofas, sleep sofas, rockers, recliners, and other chairs). For product categories for which unit shipment data were not available, we estimated unit shipments by assigning average per unit values to the Census data on value of shipments. Finally, estimates of net imports were added to shipments to estimate the total number of upholstered units sold to U.S. households. For the years in which Economic Census data are not available, shipment estimates were based on furniture shipment values published by the Department of Commerce in the Annual Survey of Manufactures.

Estimated shipments before 1967 were based on the Federal Reserve's annual furniture production index.

The CPSC's Product Population Model uses sales data and information on the average product life to estimate the numbers of items remaining in use in the years following their purchase by consumers. The estimated average useful life of upholstered furniture reportedly ranges from 15 to 17 years.

Based on the assumption that the expected life of a piece of upholstered furniture is 16 years, the average number of upholstered items in household use during 2002-2004 was about 447 million pieces.

Based on discussions between industry officials and Department of Commerce personnel.

Surveys of furniture manufacturers in the last several years show the shift towards thermoplastic fabrics peaked during the period of the mid-1980's to the mid-1990's. Information provided to the CPSC by the Upholstered Furniture Action Council (UFAC) showed that a significant shift to greater use of thermoplastic fabrics began in the 1950's, and became more pronounced in the 1970's.

These data on usage of different types of fabrics over the years can be used to characterize upholstery fabrics found on furniture in U.S. households. An estimated 31.2 percent of furniture in use in U.S. households during the period 2002-2004 was covered with fabrics predominantly made with cellulosic fabrics; an estimated 50.2 percent were covered with predominantly thermoplastic fabrics, and 18.6 percent were covered with other materials (mainly leather, wool, and vinyl-coated fabrics).

Report to the CPSC on the UFAC Voluntary Program, Upholstered Furniture Action Council, March 21, 1978.

5. Expected Benefits of the Proposed Standard

The expected benefits of the proposed standard are estimated as the reduction in the societal costs associated with upholstered furniture fires that would be prevented by the standard. We estimate the benefits in several steps. First, the average annual societal costs of upholstered furniture fires are estimated, based on estimates of the aggregate annual costs of fire-related deaths, injuries, and property damage. These costs are differentiated by ignition source (i.e., cigarette vs. open flame ignition) and by fabric covering type (since different fabrics exhibit different ignition propensities). Societal costs are also estimated on a “per product in use” basis, based on

estimates of the numbers of furniture items in use.

Second, since each furniture item is expected to remain in use for an average of 15 to 17 years, the present value of the product's estimated lifetime fire costs is estimated by summing the discounted annual costs over the item's expected useful life. The estimated annual societal costs that are expected to accrue over the furniture item's useful life are discounted at an annual rate of 3 percent. This rate is consistent with recommendations in the economic literature for discounting the costs and consequences of health programs.

Societal costs have also been estimated using a 7 percent discount rate, as recommended by the Office of Management and Budget (in addition to 3 percent) in its guidance to Federal agencies on the use of discounting in regulatory analysis (Circular A-4).

For example: Viscusi, W.K., “Discounting Health Effects for Medical Decisions,” in

Valuing Health Care: Costs, Benefits, and Effectiveness of Pharmaceuticals and Medical Technologies,

ed. F.A. Sloan, 123-24. New York: Cambridge University Press. 1995. Also, Gold, Marthe R., et al.,

Cost-Effectiveness in Health and Medicine.

New York: Oxford University Press. 1996.

Third, the expected effectiveness of the proposed standard (i.e., the percentage reduction in fire losses) is estimated for each ignition source and upholstery cover type. As discussed below, effectiveness of the standard at reducing societal costs is based on judgments regarding improvements attributed to fabric treatments and effectiveness of barrier materials.

We begin the analysis by evaluating the societal costs of cigarette fires and the expected benefits associated with preventing these fires. This is followed with an evaluation of the societal costs and likely benefits associated with the prevention of open-flame ignited fires.

a. Expected Benefits From Reducing Cigarette Fire Losses

Societal costs of furniture fires started by cigarettes.

The purpose of this section is to estimate the societal costs of cigarette-related upholstered furniture fires to use as the basis for estimating the cigarette benefits. In the next section, benefits are estimated as avoided societal costs. These costs are based on fire losses (deaths, injuries and property loss) estimated by the CPSC Directorate for Epidemiology, which relies on fire loss data acquired from the National Fire Protection (NFPA) annual survey of fire departments and the U.S. Fire Administration (USFA) National Fire Incident Reporting System (NFIRS). The most recent fire data available to make such estimates was for the 2002-2004 time period. Societal cost estimates are also differentiated by fabric cover types, which (as described below) exhibit different cigarette ignition propensities.

According to the CPSC's Directorate for Epidemiology, there was an average of 260 addressable civilian deaths and 320 nonfatal civilian injuries annually from fires started by cigarettes during the 2002-2004 time frame.

There was also an average of about $73 million annually (in 2005 dollars) in property losses from cigarette-ignited fires.

By combining the costs associated with deaths, injuries, and property damage, total societal costs can be estimated.

Miller, David. “2002-2004 Fire Loss Estimates for Upholstered Furniture.” Directorate for Epidemiology, U.S. Consumer Product Safety Commission, August 3, 2007 (Draft). The Directorate for Epidemiolgy based its estimates on a methodology that was refined to address concerns raised by the General Accounting Office (GAO) in a 1999 report, “Consumer Product Safety Commission: Additional Steps Needed to Assess Fire Hazards of Upholstered Furniture.”

Estimated average property losses of about $65 million for 2002-2004 (Miller,

op. cit.

) are expressed in 2004 dollars ($70 million) based on changes in the Producer Price Index for construction materials.

For analytic purposes staff assigns a value of $5 million as the value of a statistical life for the calculation of societal costs. The $5 million estimate is consistent with the general range of the value of a statistical life published in the literature, which generally falls in the $3 million to $7 million range.

Multiplying the annual estimate of about 260 deaths by the value of a statistical life of $5 million yields annual fatality costs of $1.3 billion.

Viscusi, W. Kip, “The Value of Risks to Life and Health,” Journal of Economic Literature, Vol. XXXI, December 1993, pp. 1912-1946.

Nonfatal injuries were assigned an average cost of $146,740 each. The basis for this estimate was the analysis of burn injury costs reported in the August 1993 report “Societal Costs of Cigarette Fires,” part of the research sponsored by the CPSC under the Fire Safe Cigarette Act of 1990.

44 45

The $146,740 figure represents a weighted average of injury costs (including pain and suffering) for both hospitalized injuries and injuries treated and released. The estimate of 320 injuries annually results in societal costs of about $47 million.

Zamula, William W., “Costs for Non-Fatal, Addressable Residential Civilian Injuries Associated with Upholstered Furniture Fires.” (Memorandum to Gregory B. Rodgers, AED, EC) Directorate for Economic Analysis, U.S. Consumer Product Safety Commission. September 6, 2007. (Costs are estimated in 2005 dollars.)

Miller, Ted R., et al., “Societal Costs of Cigarettes Fires,” prepared for the U.S. Consumer Product Safety Commission under the Cigarette Safety Act of 1984, August 1993.

As noted above, the proposed standard would also address about $70 million annually in property losses from fires started by cigarettes, based on estimates for the 2002-2004 period. Consequently, the total annual costs of cigarette-ignited fires addressed by the proposed standard amounted to an annual average of about $1,420 million ($1,300 million + $47 million + $73 million) during the 2002-2004 time period.

Information on the number of furniture items (i.e., separate pieces of furniture) in use provides a basis for estimating the costs of cigarette ignition fires on a per unit basis. The average estimated number of items of residential living room and family room upholstered furniture in use during the 2002-2004 time period was about 447 million units, based on an expected useful product life of 15-17 years. Given the annual societal costs and the number of furniture units in use, the annual societal cost per unit of furniture in use, resulting from cigarette ignition, amounted to about $3.18 ($1,420 million/447 million units of furniture). This per unit societal cost estimate represents an average across all furniture items in use. However, because different fabric coverings for furniture exhibit different ignition propensities, we can develop more precise estimates of per unit societal costs by accounting for the fabric cover.

Ignition testing of chairs by CPSC staff and others over the years has shown that the cigarette ignition hazard of furniture mainly involves chairs covered with fabrics that are predominantly woven from cellulosic fibers, i.e., cotton and rayon. Chair testing done by the CPSC staff and California's Bureau of Home Furnishings has shown that chairs covered with predominantly thermoplastic fabrics (e.g., polyester, polypropylene, and nylon) are much less likely to ignite from cigarettes. Chairs covered with some materials, such as leather, vinyl-coated fabrics, and wool fabrics are resistant to ignition from cigarettes. Given the disparity of ignition propensities, some types of furniture would be expected to result in greater societal costs from fires. Information relevant to the determination of average ignitability and estimation of societal costs for furniture covered with different types of materials is discussed below.

The results of the analysis described in this section (including estimates of market shares by fabric covering, estimates of ignition propensities and risk by fabric type, and estimates of

annual societal costs) are summarized in Table 1.

BILLING CODE 6355-01-P

EP04MR08.000

BILLING CODE 6355-01-C

Estimates of the types of upholstery on furniture pieces found in households during 2002-2004 were derived from historical data from surveys in various years, estimates of annual sales of upholstered furniture, and calculations of the survival of furniture in years after purchase (using the CPSC's Product Population Model). Based on these sources, the Directorate for Economic Analysis estimates that 50.2 percent of the 447 million upholstered furniture items that were in use during 2002-2004 were covered with thermoplastic fabrics, 31.2 percent were covered with cellulosic fabrics, and 18.6 percent were covered with leather, vinyl-coated fabrics, or wool fabrics. These market shares are shown in Table 1, column 1.

Note that the market shares in the first three rows sum to the 31.2 percent of the furniture in use covered with cellulosic fabrics. However, because extensive testing data show that some cellulosic fabrics are more likely to ignite than others, this analysis also separates cellulosic fabrics into three categories according to their ignition propensities. The next several paragraphs describe this sub-categorization of cellulosic fabric coverings.

Testing by the CPSC laboratory using the proposed Upholstery Fabric Smoldering Ignition Test

indicates that upholstery cover materials which are most likely to fail the test are fabrics woven entirely of cellulosic fibers that are heavier than eight ounces per square yard. These fabrics are assumed to include all fabrics that would be classified as Class II fabrics under the UFAC Program as well as predominantly cellulosic fabrics that would be classified as Class I fabrics under the UFAC Program and Class C and D fabrics according to the proposed furniture flammability standard fabric test method developed by the National Bureau of Standards (NBS, now the National Institute of Standards and Technology) in the 1970s. Estimation of the percentage of fabrics that would fail the fabric test of the proposed standard, and assessment of the societal costs presented by different types of upholstery cover materials are, therefore, based on fabric and chair test data accumulated over the years.

The Upholstery Fabric Smoldering Ignition Test

is cigarette ignition testing of fabrics over a standard non-flame-retardant polyurethane foam substrate.

Classification of cellulosic fabrics according to the test developed by UFAC (which classifies fabrics according to char length on the vertical surface when tested over standard non-FR polyurethane foam) and the test developed by NBS (which classifies fabrics according to char length when tested over a glass fiberboard substrate) have been used to categorize the ignition performance of cellulosic fabrics in this analysis. CPSC laboratory analyses since 1980 found that about 82 percent of cellulosic fabrics tested were Class I fabrics according to the fabric classification test of the UFAC Program (i.e., having a vertical char length of less than 1.75 inches), and 18 percent of cellulosic fabrics were UFAC Class II fabrics (i.e., having a vertical char length of 1.75 inches or greater). Assuming the tested fabrics were representative of cellulosic fabrics, 25.6 percent of all fabrics on furniture in use during 2002-2004 were UFAC Class I (31.2% that were covered with cellulosic fabrics × 82%) and 5.6 percent were UFAC Class II (31.2% × 18%).

Laboratory testing shows that the cover material smoldering resistance test of the proposed standard is more severe than the UFAC Fabric Classification Test.

Therefore, for the purposes of this analysis, UFAC Class II fabrics are assumed to fail the proposed fabric test without changes that would improve their ignition resistance. Limited testing also indicates that some portion of UFAC Class I fabrics will fail the fabric test of the proposed standard. Twenty-five percent of the Class I fabrics tested by the CPSC staff in 1980 and 1984 were found to be generally more ignition-prone Class D fabrics according to the NBS fabric classification test (i.e., sustaining chars of greater than 3 inches when tested over glass fiberboard). If we assume that such fabrics would fail the proposed standard's fabric test, an estimated 12 percent of fabrics found on furniture in 2002-2004 would have failed the test (5.6 percent which were UFAC Class II, plus 25 percent of the 25.6 percent of other cellulosic fabrics which were UFAC Class I. (Designated as “Severely Ignition-Prone Cellulosics” in Table 1.)

Tao, Weiying, Ph.D. “Evaluation of Test Method and Performance Criteria for Cigarette Ignition (Smoldering) Resistance of Upholstered Furniture Materials.” Division of Electrical and Flammability Engineering, Directorate for Laboratory Sciences, U.S. Consumer Product Safety Commission. May 2005.

Fabrics assumed to pass the proposed standard include more moderately ignition-prone fabrics that are Class I according to the UFAC Fabric Classification test and Class C according to the NBS fabric test (i.e., sustaining chars of 1.5—3 inches when tested over glass fiberboard), and more ignition-resistant Class B cellulosic fabrics according to the NBS fabric test (which sustain char lengths of less than 1.5 inches when tested over glass fiberboard). The Class C fabrics accounted for an estimated 5.8 percent of fabrics found on furniture in 2002-2004 (22.5 percent of UFAC Class I cellulosic fabrics according to CPSC staff testing). These fabrics are designated as “Moderately Ignition-Prone Cellulosics” in Table 1. More ignition-resistant NBS Class B fabrics are estimated to have comprised 52.5 percent of UFAC Class I cellulosic fabrics, or 13.4 percent of all fabrics and covering materials found on upholstered items in 2002-2004. These fabrics are designated as “Lower Ignition-Prone Cellulosics” in Table 1.

Estimated ignition propensities for furniture covered with cellulosic fabrics are based on chair testing that was done in 1984 and 1994. Evaluating chair test results according to UFAC and NBS fabric classifications, 58.3 percent of test cigarettes were estimated to lead to ignitions for chairs covered with UFAC Class II fabrics. The estimated ignition propensity for test cigarettes on chairs covered with UFAC Class I, NBS Class D fabrics was 46.6 percent. Combining these two severely-ignition-prone fabric classes yields an average estimated ignition propensity of 52.1 percent (weighted by their 2002-2004 market shares). Cigarettes placed on furniture covered with moderately ignition-prone fabrics had an estimated 32.2 percent likelihood of resulting in ignition.

About 10.5 percent of test cigarettes were estimated to lead to ignitions for chairs covered with less ignition-prone cellulosic fabrics.

(See column 2 of Table 1.)

UFAC Class I, NBS Class C cellulosic fabrics.

NBS Class B cellulosic fabrics.

Because of less concern with the ignition propensity of thermoplastic fabrics, ignition testing data for such materials are more limited. Expanding chair test data to include tests conducted in 1980 led to an estimate that 1.5 percent of test cigarettes would result in ignition for furniture covered with thermoplastic fabrics. Additionally, based on limited laboratory ignition testing data, materials such as leather, wool fabrics, and vinyl-coated fabrics are assumed to be highly resistant to ignition from cigarettes.

The calculation of weighted ignition propensities of furniture covered with different types of fabrics is the product of the estimated market share of furniture in use in 2002-2004 for each type of fabric and its estimated ignition propensity. The estimated weighted ignition propensity was 0.063 for items covered with severely ignition-prone

cellulosic fabrics (i.e., 12.0% share of the market × 52.1% ignition propensity); 0.019 for items covered with moderately ignition-prone cellulosic fabrics (5.8% × 32.2%); 0.014 for items covered with less ignition-prone cellulosic fabrics (13.4% × 10.5%); and .008 for items covered with thermoplastic fabrics (50.2% × 1.5%). (See column 3 of Table 1.)

The percent of total risk presented by furniture covered with different fabric types was derived by dividing estimated weighted ignition propensities by the sum of all weighted ignition propensities (which was about .103 for furniture in use in 2002-2004). Thus, as shown in the table, the more severely ignition-prone cellulosic fabrics

were estimated to account for 60.9 percent of the total risk (.063/.103); moderately ignition-prone cellulosic fabrics

accounted for an estimated 18.0 percent of the risk (.019/.103); less ignition-prone cellulosic fabrics accounted for about 13.7 percent of the risk (.014/.103); and thermoplastic fabrics accounted for about 7.3 percent of the risk (.008/.103). (See column 4 of Table 1.)

UFAC Class II and UFAC Class I/NBS Class D fabrics.

NBS Class C cellulosic fabrics.

Percent of total risk for each fabric type was calculated from estimates of market share and ignition propensity that were not rounded.

The average annual societal costs associated with cigarette ignitions of each fabric type were estimated by dividing the product of estimated percent of total risk (above) and the total estimated average annual societal costs associated with cigarette ignition of furniture ($1,420 million) by the estimated number of units in use during 2002-2004 with each fabric type (447 million units in use × estimated market share). The average annual societal costs were estimated to be $16.08 for items covered with severely ignition-prone cellulosic fabrics (60.9% × $1,420 million/447 million × 12.0%); $9.94 for items covered with moderately ignition-prone cellulosic fabrics (18.0% × $1,420 million/447 million × 5.8%); $3.24 for items covered with less ignition-prone cellulosic fabrics (13.4% × $1,420 million/447 million × 13.7%); and $.46 for items covered with thermoplastic fabrics (7.3% × $1,420 million/447 million × 50.2%). (See column 5 of Table 1.)

The estimated lifetime societal costs per unit of furniture were calculated as the present value of the estimated annual societal costs over the expected product life of the item of furniture. The annual expected societal costs of cigarette ignition were assumed to apply each year that an item of furniture remains in household use. The CPSC's Product Population Model was used to calculate the likelihood that furniture items would remain in use in years after purchase. Annual societal costs per unit were multiplied by estimated probability of survival in subsequent years. The estimated stream of future expected societal costs were discounted to their present values, using a discount rate of 3 percent.

Available data suggest that other factors (in addition to changes in fabrics) have contributed to a decline in fires resulting from cigarette ignition of upholstered furniture over time. These factors include changes in smoking-related behavior of individuals, increased presence of smoke alarms, and changes in furniture filling materials. The present value estimates were further adjusted to account for an expected future decline in smoking-related fire incidents. This was done by forecasting future fire deaths by year, based on trends in deaths from cigarette ignitions of upholstered furniture during 1980-2004, and reducing the expected societal costs of cigarette ignited fires by the projected percentage reduction. This analysis found that expected lifetime societal costs, discounted to their present value using a 3 percent discount rate, should be reduced by approximately 28 percent. Thus, expected lifetime societal costs per unit of $195.31 for items covered with severely ignition-prone cellulosic fabrics were reduced to $140.04 after incorporating the trend data. Similar calculations led to estimates of lifetime societal costs of $86.60 for items covered with moderately ignition-prone cellulosic fabrics; $28.24 for items covered with less ignition-prone cellulosic fabrics; and $4.06 for items covered with thermoplastic fabrics. (See column 6 in Table 1.)

b. Expected Benefits

The analysis described above estimated the per unit hazard costs associated with the upholstery materials of different ignition propensities, based on the furniture in use during 2002-2004, the most recent time period for which fire data is available. However, as discussed in Section 4, the types of upholstery materials used in the production of furniture have changed over the years. Since the proposed standard would address risks associated with current production, projection of benefits requires estimating the societal costs associated with materials now being used to manufacture furniture. This is accomplished by estimating the percentage of furniture items currently made with covering materials of differing ignition propensities.

A 2006 survey of furniture manufacturers by Ciprus Limited provides information on consumption of cellulosic, thermoplastic, and leather covering materials in the production of furniture.

Using CPSC staff test data discussed above, the percentages of current production (as indicated by the Ciprus data) made with materials ranging from severely ignition-prone cellulosic fabrics to ignition resistant materials such as leather were estimated. These estimates are shown in column 1 of Table 2. The estimated percentage of upholstered items now made with severely ignition-prone cellulosic fabrics has fallen to 9.6 percent of annual production, from 12.0 percent estimated for furniture in use during 2002-2004. This is a 20 percent decrease in the relative use of the most ignition-prone class of fabrics. The use of other ignition-prone fabrics has also declined, in relative terms, while the use of generally ignition-resistant materials such as leather (estimated to be about 30 percent of current production) is 62 percent greater than found in household use in 2002-2004.

Ciprus Limited,

op. cit.

BILLING CODE 6355-01-P

EP04MR08.001

BILLING CODE 6355-01-C

Column 2 of Table 2 shows the expected number of furniture units produced annually, by type of covering material, based on the market shares of the various fabric coverings (column 1) and an estimated 30.5 million furniture units produced. Column 3 provides the estimates of per unit lifetime societal costs derived in Table 1.

Based on current estimates of the types and quantity of furniture produced, the estimated total present value of the expected societal costs from cigarette fires is $681 million for furniture produced in a year, in the absence of a standard. (See column 4 of Table 2.) Total estimated societal costs involving furniture covered with severely ignition-prone cellulosic fabrics account for $411 million, or about 60 percent of the total. In contrast, thermoplastic fabrics, which are used to cover about 45 percent of all upholstered furniture produced, account for an estimated $55.5 million in societal costs, or only about 8 percent of the total.

A comparison of the ignition performance of upholstered chairs made with current fabrics with that of chairs made in compliance with the proposed standard would provide data to assess the likely reduction in ignition propensity that would result from the proposed standard. In the absence of such data, we can estimate the benefits of the standard by making reasonable judgments about improvements in ignition performance that would result from the use of complying materials.

Furniture currently manufactured with severely ignition-prone cellulosic fabrics could realize a reduction in societal costs per unit under the proposed standard to the equivalent of that now estimated for furniture covered by less ignition-prone cellulosic fabrics. This reduction would be attributable to improved ignition performance of fabrics or from the use of qualifying barriers. The reduction in lifetime societal costs per unit from $140.04 to $28.24 amounts to a hazard reduction of 79.8 percent (shown in column 5 of Table 2). We likewise assume that pre-standard societal costs estimated for moderately ignition-prone cellulosic fabrics (which are also expected to fail the proposed cover fabric test) would also likely fall to the level of estimated hazard costs associated with furniture covered with less ignition-prone fabrics. The estimated reduction from estimated lifetime societal costs of $86.60 to $28.24 would be a 67.4 percent reduction in the hazard presented (also shown in column 5). Since upholstered furniture items covered with less ignition-prone cellulosic fabrics and thermoplastic fabrics are expected to pass the proposed cover fabric test, and there are no requirements for filing materials under the proposed standard, furniture covered with those fabrics would not be expected to be associated with any reduction in their expected societal costs.

The estimated benefits per unit were calculated for each fabric class. (See column 6 of Table 2.) Per unit benefits of the proposed standard range from $0 for furniture covered with ignition-resistant fabrics such as thermoplastic or lower cigarette-ignition-prone cellulosics to an estimated $111.80 per unit for items currently covered by severely ignition-prone cellulosic fabrics. The benefits from ignition resistant materials such as leather, wool, and vinyl-coated fabrics are also expected to be $0.

The total estimated benefits of the proposed standard are calculated by multiplying estimated per unit benefits (shown in column 6) by the estimated annual units produced with each class of covering material (column 2). Based on these calculations, estimated benefits of the standard, in the form of expected lifetime reduction in societal costs associated with production of furniture in one year, discounted to their present value using a discount rate of 3 percent, total $410.2 million. About 80 percent of total estimated benefits are associated with the approximately 10 percent of furniture currently made with severely ignition-prone cellulosic fabrics.

As noted previously, OMB guidance to Federal agencies on the use of discounting in regulatory analysis recommends that future benefits (and costs) of federal regulations be presented using discount rates of 3 percent and 7 percent. Projected benefits from reductions in smoldering ignitions have an estimated present value of $309.1 million if future benefits are discounted at a 7% discount rate.

In addition to cigarette losses, the Directorate for Epidemiology estimated fire losses from small open-flame ignitions for the years 2002-2004.

During this time period, there were an average of 30 deaths and 170 nonfatal injuries annually from fires started by small open flames. There was also an average of about $50 million annually in property losses from small open flame-ignited fires during this time frame.

Miller, David.

op. cit.

Estimated average property losses for 2002-2004 are expressed in 2005 dollars, based on changes in the Producer Price Index for construction materials.

Assuming a value of statistical life of $5 million,

the societal costs associated with the 30 deaths annually amounted to about $150 million. The 170 nonfatal injuries were assigned an average cost of $146,740 each,

resulting in societal costs of about $25 million. Adding in the $50 million annually in property losses from fires started from small open-flame ignition, the total annual costs of open-flame ignited fires addressed by the proposed standard amount to about $225 million ($150 million + $25 million + $50 million).

Viscusi, W. Kip,

op. cit.

Zamula, William W.,

op. cit.

Injury costs are expressed in 2005 dollars.

As in Table 1, these annual estimates of the open-flame losses are used to develop estimates of the lifetime societal costs of open-flame hazards per unit of furniture in use during 2002-2004, for each of the five fabric categories. The results are presented in Table 3.

BILLING CODE 6355-01-P

EP04MR08.002

BILLING CODE 6355-01-C

Column 1 of Table 3 shows the proportions of furniture in each fabric material category, and is identical to the corresponding column in Table 1. Column 2 describes open-flame ignition propensities, based on small open flame ignition testing by the CPSC laboratory in 1996. In that testing, cellulosic and thermoplastic fabrics had nearly the same ignition propensity when subjected to a small flame for 20 seconds. Ignitions in 20 seconds or less were observed for 27 of 29 predominantly cellulosic fabrics (about 93 percent) and 17 of 18 predominantly thermoplastic fabrics (about 94 percent).

Based on testing data presented in Directorate for Laboratory Sciences memoranda dated October 3, 1996, through September 19, 1997, Tab D, “Upholstered Furniture Flammability: Regulatory Options for Small Open Flame & Smoking Material Ignited Fires,” October 24, 1997.

Based on these ignition propensities and the estimated percentages of furniture in use comprised by upholstered items with cellulosic and thermoplastic fabrics, furniture covered with thermoplastic fabrics accounted for an estimated 62 percent of the overall risk of small open flame ignitions during 2002-2004; items covered with cellulosic fabrics accounted for an estimated 38 percent of the risk. While Table 3 separates cellulosic fabrics according to differences in their cigarette ignition propensities, for this analysis all cellulosic fabrics are assumed to have the same small open flame ignition propensity. The estimated percent of overall risk for each type of cellulosic fabric is, therefore, determined by market share. As with the risk of ignition by cigarettes, furniture covered by leather, wool, and vinyl-coated fabrics is assumed to be resistant to ignition from a 20-second exposure to a small open flame.

Following the same methodology described in Table 1, the average annual societal costs associated with small open flame ignitions of each fabric type were estimated by dividing the products of estimated percent of total risk and the total estimated average annual societal costs associated with small open flame ignition of furniture ($225 million) by the estimated number of units in use during 2002-2004 with each fabric type (447 million units in use × estimated market share). This approach resulted in estimated average annual societal costs of about $.62 for items covered with thermoplastic fabrics (62% × $225 million /447 million × 50.2%) and about $.61 for items covered with predominantly cellulosic fabrics (38% × $225 million/447 million × 31.2%). (See column 5 of Table 3.)

Finally, the lifetime societal costs (per unit of furniture) were estimated as the present value of the annual per unit societal costs over the expected product life of a furniture item. This present value estimate (shown in column 6), discounted at a rate of 3 percent, is about $7.55 for items covered with predominantly thermoplastic fabrics and $7.44 for items covered with predominantly cellulosic fabrics.

The estimated benefits associated with the prevention of open-flame fires are described in Table 4. The methodology is similar to that described for Table 2. Column 1 shows the current market shares, by fabric type, and Column 2 shows annual sales based on annual furniture shipments of 30.5 million units. Column 3 provides the estimates of per unit lifetime societal costs derived in Table 3, and Column 4 provides estimates of the aggregate societal costs of fires associated with open-flame ignition.

BILLING CODE 6355-01-P

EP04MR08.003

BILLING CODE 6355-01-C

For the purposes of this analysis, we assume that about 40 percent of furniture currently manufactured with severely cigarette ignition-prone cellulosic fabrics (accounting for about 1.17 million units, or 3.8 percent of all furniture items) would be made with barrier materials. Complying barriers may reduce the open flame ignition hazards by about 90 percent, or $6.70 per unit, and benefits could total $7.9 million for furniture made with complying barriers.

Based on the assumption that 40 percent of severely cigarette ignition-prone cellulosic fabrics would be used with complying barriers, the remaining 60 percent of furniture currently manufactured with severely cigarette ignition-prone cellulosic fabrics (accounting for 5.8 percent of all furniture items) and the 4.6 percent of fabric yardage that is moderately cigarette ignition prone (combining for nearly 3.2 million units) would require other modifications or they would have to be dropped from use as upholstery cover materials. The methods of compliance chosen by manufacturers likely would affect the level of reduction in open flame ignition hazards. The implications of these decisions are discussed below.

Fabrics that do not pass the upholstery cover fabric smoldering ignition resistance test could be brought into compliance through treatments with FR chemicals. FR treatment of fabrics and filling materials to achieve compliance with the staff's 2005 draft standard might result in a 50 percent reduction in small open flame fire losses.

However, unlike the 2005 draft standard, the current proposed standard does not include provisions related to open flame ignition performance of filling materials, which in many cases would have required FR treatments to achieve compliance. Lacking this additional contribution to fire-retardance, the effectiveness of FR fabric treatments under the proposed standard at reducing the small open flame fire hazard probably would be lower. Consequently, the hazard reduction for furniture with FR-treated fabrics may be about 25 percent under the proposed standard. Per unit open flame ignition benefits would be about $1.86, and aggregate open flame benefits would be about $5.9 million, if manufacturers resort to FR treatment for all of the nearly 3.2 million units. From the standpoint of fabric type, the average hazard reduction for severely cigarette ignition-prone cellulosic fabrics would be 51 percent,

and the reduction for moderately cigarette ignition-prone cellulosic fabrics would be 25 percent. (See column 5 of Table 4.)

Smith, Charles, Directorate for Economic Analysis, CPSC, Preliminary Regulatory

Analysis of a Draft Proposed Flammability Rule to Address Ignitions of Upholstered Furniture,

November 2007.

Based on 25% effectiveness x 60% of the fabrics being FR-treated and 90% x 40% that are made with barriers.

Alternatively, manufacturers would have the options of using fabrics that are reformulated with different fibers or dropping non-complying fabrics from use as furniture covers. In fact, this may be the preferred option for most manufacturers, given concerns with costs, FR exposure, aesthetic effects, and other issues. Open flame benefits would not be expected for such furniture items. If the use of FR-treatments of fabrics is 80 percent lower than assumed above, the number of units made with FR-treated fabrics would total about 630,000 and aggregate open flame benefits from furniture using FR-treated fabrics would be about $1.2 million, and total open flame benefits would be about $9 million. If all 630,000 units with FR fabric treatments involved severely cigarette ignition-prone fabrics, the average estimated hazard reduction for that category of fabrics would be about 41 percent.

Based on 25% effectiveness x 21.6% of the fabrics being FR-treated and 90% x 40% that are made with barriers.

Based on the assumed range of furniture units that would be made with FR-treated fabrics, aggregate open flame benefits from the proposed standard range from about $9 million to $13.8 million, as shown in column 7 of Table 4. In accordance with OMB guidance that future benefits (and costs) of federal regulations be presented using discount rates of 3 percent and 7 percent, open flame benefits of the proposed standard have also been estimated to have a present value of $6.4 million to $9.9 million if future benefits are discounted at a 7 percent discount rate.

6. Expected Costs of the Proposed Standard

a. Costs Related to Upholstery Fabrics and Barrier Materials

Upholstery fabric and FR treatments.

This section of the analysis presents information about the expected resource costs associated with the proposed standard. These costs include manufacturing costs incurred for materials, labor, testing, and recordkeeping, and distribution costs to wholesalers, distributors, and retailers. The estimates are expressed in 2005 dollars (as were estimated benefits). Cost estimates are limited to upholstered household furniture that may commonly be found in living rooms and family rooms. A relatively small number of other types of chairs that fall within the scope of the standard, such as a small percentage of dining chairs and desk chairs purchased by consumers, are excluded from this analysis.

Cost estimates are summarized in Table 5.

Those other items probably would incur relatively minor increases in costs because of the types of materials used, and smaller material requirements per unit of furniture.

BILLING CODE 6355-01-P

EP04MR08.004

BILLING CODE 6355-01-C

Fabrics failing the fabric test of the proposed standard could be treated with FR chemicals or be reformulated with fibers that enable passing results. Manufacturers would also be able to continue using fabrics without modifications if they use an acceptable barrier material (i.e., one that passes the proposed barrier tests) between the fabric and filling materials. For purposes of this analysis, the highly cigarette ignition-prone fabrics and moderately cigarette ignition-prone fabrics, estimated to combine for about 14.2 percent of total upholstery cover materials, are assumed to require modifications if their use is to continue under the proposed standard. As discussed previously, these modifications could include the use of FR treatments or barriers, or reformulating the fabrics in a way (such as increasing the thermoplastic fiber content) that will allow the fabrics to pass the smoldering test of the proposed standard.

Based on fabrics that have been tested by the CPSC laboratory, many of the fabrics that would fail the fabric test of the proposed standard are heavier weight (over eight ounces per square yard) fabrics that are made entirely of cellulosic fibers, such as cotton or rayon. Many of these fabrics could be treated with FR chemicals to enable them to pass the fabric test. Typically, fully upholstered chairs require about 7 linear yards of fabric, and sofas require 11 to 15 yards, depending on factors such as the need to match patterns (which results in more fabric waste in pattern cutting). The average increase in fabric costs could range from $.62 to $1.05 per linear yard for manufacturers, based on previous estimates for FR backcoating to achieve resistance to ignition from small open flames.

Also, although the proposed standard does not specify frequency of testing to assure compliance of treated fabrics with the fabric test, we assume that testing will be done to provide guaranties to furniture manufacturers. This testing could increase fabric costs an additional $.03 to $.06 per linear yard of fabric, on average. Therefore, total average manufacturing cost increases for furniture made with FR-treated upholstery fabrics under the proposed standard could range from $4.55 to $7.77 for chairs and $8.45 to $14.43 for sofas and loveseats.

Considering estimates of unit shipments of chairs and sofas (based on an analysis of Department of Commerce Economic Census data), the average manufacturing cost increase per item of furniture resulting from FR treatments of fabric is estimated to range from $6.61 to $11.28.

(See column 1 of Table 5.)

Smith, Charles. Directorate for Economic Analysis, CPSC,

Economic Analysis of Regulatory Options to Address Small Open Flame Ignitions of Upholstered Furniture,

October 2001. Note: Bureau of Labor Statistics reports virtually no change in Producer Price Index for job or commission finishing of cotton broadwoven fabrics from 2001-2005. Therefore, previous estimates are used in this analysis.

Assuming average fabric yardage for sofas and loveseats is 13 linear yards.

We estimate that in 1997, upholstered living room and family rooms furniture purchased for consumer use was comprised of about 15.6 million sofas, sofabeds, and loveseats (52.7%), and 14.0 million chairs (47.3%). Therefore: ($4.55 × 47.3%) + ($8.45 × 52.7%) = $6.61; and ($7.77 × 47.3%) + ($14.43 × 52.7%) = $11.28.

Barrier materials.

Some furniture manufacturers may choose to offer fabrics that do not pass the fabric classification test by using an acceptable barrier material under the cover fabric. Based on barriers used in the UK to comply with the barrier test of that country's furniture flammability standard, the cost to manufacturers could range from $2.00 to $2.47 per linear yard (reportedly 54 to 59 inches in width) for standard FR barriers, and about $2.67 to $2.94 per linear yard for down-proof barriers (i.e. having yarns and weaves suitable for encasing down).

As with FR-treated cover fabrics, testing would be done to assure compliance with the barrier test of the proposed standard. However, given expected large production runs of barriers and the greater degree of uniformity of barrier materials compared to cover fabrics, additional testing costs to furniture manufacturers could be about $.01 per yard of barrier fabric.

Smith, Charles.

op. cit.

The decision to use barriers as a means to comply with the standard is more likely to be taken by firms that serve the upper-end furniture market. These furniture items are more likely to be manufactured with interior fabrics between the cushioning materials and the upholstery covers. In a 1995 survey of furniture manufacturers, the CPSC found that about one-third of the seat, arm and back cushions were made with interior fabrics. Interior fabrics were used in an average of about 50 percent of cushions made by smaller firms, which are more likely to serve the upper-end market. To the extent that manufacturers already enclose filling materials in interliner fabrics, the FR barriers could be replacing untreated materials.

Cushions are usually purchased from fabricators that make them to the specifications of the furniture manufacturers. For seat cushions, the barrier alternative would result in a change in the interior fabric used by the cushion fabricators. For such items, barrier costs would be offset by the costs of the untreated materials, about $.30 per yard for standard interliner fabrics and $.80 per yard for down-proof interliner fabrics. Net increases in material costs, including costs for testing, would be about $1.71 to $2.18 per yard for standard fabrics and $1.88 to $2.15 per yard for down-proof fabrics. Cushions typically have sides that are about 24 inches long, and they are about 5 inches thick. Therefore, about one linear yard of 54-inch wide interior fabric would be used per seat cushion, and the cost increases per linear yard of material would also hold true for cost increases per cushion.

Barrier materials required for other parts of the seating areas of furniture items might require about two yards of material per chair and four yards per sofa. These areas may be less likely to have interliner fabrics currently than is the case with seat cushions. Therefore, increased material costs probably would be $2.01 to $2.48 per linear yard for standard FR barriers. These materials would increase material costs by about $4.02 to $4.96 for chairs and $8.04 to $9.92 for sofas. Adding the approximately $1.71 to $2.18 per cushion material cost increases from substituting the use of FR barriers for standard interliner materials, total increased material costs might be about $5.73 to $7.14 for chairs and $13.17 to $16.46 for sofas.

In addition to increased material costs, manufacturers would also be faced with additional costs related to labor needed to include FR barriers on parts of the upholstered items that are not currently made with interliner fabrics or battings. The additional labor required might average about 15 to 20 minutes per item.

Hourly labor costs, including benefits, are estimated to range from about $25 to $30.

Therefore, labor costs for the additional upholstery work could be about $6.25 to $10.00. Total increases in

manufacturing costs (material and labor) are estimated to range from $11.98 to $17.14 for chairs and $19.42 to $26.46 for sofas and loveseats. The average increase in manufacturing costs per item of upholstered furniture that would be made with FR barriers is estimated to range from $15.90 to $22.05.

(See column 2 of Table 5.)

Based on a telephone conversation between a representative of Vanguard Furniture, and Charles Smith, Directorate for Economic Analysis, CPSC, on February 23, 2001.

Although the Bureau of Labor Statistics National Compensation Survey reports that average upholsterer wages for the Hickory-Morganton-Lenoir, NC area were $17.03 per hour in 2005, we assume that wages and other labor costs are typically higher ($25-$30) for upholsterers that work for manufacturers using expensive decorative fabrics (which are more likely to be used with barrier materials). This assumption is supported by labor cost information provided by Vanguard Furniture, op. cit.

We estimate that in 1997, upholstered living room and family rooms furniture purchased for consumer use was comprised of about 15.6 million sofas, sofabeds, and loveseats (52.7%), and 14.0 million chairs (47.3%). Therefore: ($11.98 × 47.3%) + ($19.42 × 52.7%) = $15.90; and ($17.14 × 47.3%) + ($26.46 × 52.7%) = $22.05.

As noted above, highly cigarette ignition-prone fabrics, estimated to comprise 9.6 percent of total upholstery cover materials, could require the use of FR treatments or barriers if their use is to continue under the proposed standard. The use of barriers is more economically feasible with more expensive fabrics, such as those produced by members of the Decorative Fabrics Association (DFA). The DFA estimates that fabrics marketed by its members comprise perhaps 1.5 percent of total upholstery fabric yardage used to make furniture.

If 40 percent of highly cigarette ignition-prone fabrics (3.8% of all upholstery cover materials, i.e., more than just the 1.5 percent of fabric yardage reportedly marketed by DFA members) are assumed to be used with acceptable barrier materials under a standard, about 1.17 million furniture pieces annually might be made with barriers under a standard. The aggregate manufacturing cost increase related to use of complying barrier fabrics under these assumptions would range from about $18.7 million to $25.9 million.

If 60 percent of highly cigarette ignition-prone fabric yardage (covering 5.8% of all furniture items) is assumed to be treated with FR chemicals, the estimated aggregate increase in manufacturing costs from FR treatment of fabrics would range from $11.6 million to $19.9 million annually.

The combined aggregate costs of fabric treatments and barriers would total $30.3 million to $45.7 million annually.

Information provided to the staff at a June 29, 2000, public meeting.

(30.5 million units × 3.8% × $15.90) = $18.7 million; (30.5 million units × 3.8% × $22.05) = $25.9 million.

(30.5 million units × 5.8% × $6.61) = $11.6 million; (30.5 million units × 5.8% × $11.28) = $19.9 million.

In addition to costs associated with furniture covered with severely cigarette ignition-prone cellulosic fabrics, fabrics that are moderately cigarette ignition-prone could also be expected to require modifications in order to comply with the proposed standard's smoldering ignition test for cover materials. If these units (accounting for an estimated 4.6% of current furniture purchases by consumers) are also made with FR fabric treatments, material costs per unit would increase by $6.61 to $11.28, for an increase in estimated aggregate costs ranging from $9.3 million to $15.9 million annually. Total estimated material cost increases related to FR treatment of fabrics or the use of complying barriers would, therefore, range from about $39.6 million to $61.6 million annually.

It should be noted that these cost estimates could be considered to be the upper bound for material costs of the proposed standard, since manufacturers would have the less expensive alternative of substituting upholstery fabrics that pass the smoldering requirements for those that do not, without the application of FR chemicals or the use of barrier materials. If choosing these options were to reduce reliance on FR-treatments of fabric by 80 percent from that assumed in the above analysis, FR-treatment costs under the proposed standard could total about $6.3 million annually. Under this assumption, an estimated 2.1 percent of furniture items would be made with FR-treated fabrics; 3.8 percent would be made with barrier materials, and; 8.3 percent would be units in which fabrics were reformulated with more ignition-resistant fibers or otherwise switched to fabrics/covers that comply without treatments or barriers. In this scenario, aggregate costs of FR-treatment of fabrics and the use of barriers would be about $30.8 million.

b. Costs Related to Compliance Verification

Costs related to compliance verification will result from requirements placed on furniture manufacturers to maintain records and to apply a permanent label to the items.

Other resource costs of compliance verification include the costs of compliance and enforcement activities undertaken by CPSC staff. For purposes of this analysis we assume compliance verification costs of about $.10 per furniture unit. (See column 5 of Table 5.)

Costs related to production testing are incorporated in the estimated material costs of the draft standard.

c. Distribution Costs

An additional cost of the proposed standard could be increases in costs to wholesalers, distributors, and retailers in the form of added storage, transportation, and inventory financing costs. Since furniture items that would be produced under the standard are not likely to be larger or heavier than pre-standard items, added storage and transportation costs are likely to be negligible. However, inventory financing costs will increase by the average cost of borrowing money, applied to the increase in the wholesale price of a furniture item over the average inventory holding time period. Since most furniture producers use just-in-time production and have small inventories of finished items, this additional cost will probably not exceed 10 percent of the increase in manufacturing costs. A 10 percent markup, therefore, is being used to measure these distribution costs. This yields a resource cost to the firms in the distribution chain averaging about $0.67-$1.14 per furniture item made with FR-treated fabrics and $1.60 to $2.22 per item made with barriers. The weighted range of estimated resource costs for furniture made with severely cigarette ignition-prone fabrics is $1.04 to $1.57 per unit of furniture.

(See column 4 of Table 5.) Aggregate costs associated with estimated increased inventory financing costs range from $4.2 million to $6.4 million annually. As discussed in Section 7 of this analysis, the proposed standard may lead to increases in retail prices of furniture greater than the 10 percent markup.

Based on the assumption that 60% of these units will use FR-treated fabrics and 40% will use barriers.

d. Summary of Expected Costs

Table 5 summarizes the results of the cost analyses. It illustrates the differing costs estimated to be incurred under the standard by furniture items covered with the different classifications of upholstery materials previously discussed in the societal costs and benefits section of this analysis. The estimated 14.2 percent of furniture items covered by severely and moderately cigarette-ignition-prone cellulosics would incur greater total and per unit costs under the proposed standard. We assume these fabrics would fail the upholstery cover fabric smoldering ignition resistance test of the proposed standard. Therefore, their continued use in furniture production would require the use of barrier materials that pass the barrier test of the proposed standard or other treatments. Furniture items covered with other types of upholstery materials should not require FR-treated fabrics or barriers. However, all units would incur minor compliance verification costs.

Based on the estimated increases in manufacturing costs associated with changes in fabrics and the use of barriers, costs of compliance verification, and distribution costs, aggregate costs under the proposed standard are estimated to range from about $47 million to $71 million annually. The midpoints of the estimated ranges of costs total $59.1 million. As noted above, since changes in fiber contents of fabrics or dropping fabrics from selections offered by manufacturers will be an option available to manufacturers, the aggregate manufacturing costs related to FR treatments and barriers could be lower. Under an alternative assumption that the reliance on FR treatments of fabrics will be 80 percent lower, aggregate costs of the proposed standard would be about $34 million for one year's production of complying furniture.

7. Comparison of Costs and Benefits

a. Benefits and Costs of Proposed Standard

The expected benefits of the proposed standard, which will vary depending on the cigarette ignition propensity of the upholstery cover material used, were discussed in Section 5 of this analysis (and shown in Tables 2 and 4) and are summarized in Table 6. Table 6 shows the estimated benefits (per unit of furniture) in columns 1, 2, and 3. The benefits associated with bringing furniture pieces now covered with severely cigarette ignition-prone cellulosic fabrics into compliance are estimated to range from $114.88 to $115.59 per unit (comprised of $111.80 from reduced losses from furniture fires started by cigarettes and $3.08 to $3.79 from reduced losses from fires started by small open flames). The projected benefits resulting from modifications to furniture covered with moderately cigarette ignition-prone cellulosic fabrics range from $58.36 to $60.22 per unit. For both groups of fabrics the range in benefits is attributable to the effect of different assumptions of use of FR fabric treatments on open flame ignition benefits. Other types of covering materials are not expected to be associated with either cigarette or open flame benefits, since no modifications to fabrics or filling materials would be required to comply with the proposed standard.

BILLING CODE 6355-01-P

EP04MR08.005

BILLING CODE 6355-01-C

Table 6 also shows (in column 4) the midpoints of the ranges of estimated per unit costs of compliance with the proposed standard, which were discussed in Section 6 of this analysis. Estimated costs per unit of furniture covered with severely and moderately cigarette ignition-prone cellulosic fabrics are expressed as ranges based on different assumptions of the extent to which FR treatment would be used to achieve compliance. The higher cost estimates reflect the midpoint of costs estimated using an assumption that all of the affected fabrics are either FR treated or used with complying barriers. The lower cost estimates assume that reliance on FR treatments is reduced by 80 percent, as manufacturers comply through fabric fiber reformulation or dropping noncomplying fabrics from use as upholstery covers.

Table 6 also shows aggregate and cumulative net benefits associated with the proposed standard. The total net benefits shown in column 7 are the product of per unit net benefits and number of units produced annually by type of cover material. For example, the total estimated net benefits from furniture covered with moderately cigarette ignition-prone cellulosic fabrics range from $70.7 million to $81.9 million, given by the product of 1.4 million units produced and per unit net benefits of $50.27 to $58.25. The cumulative net benefits (shown in column 8 of Table 6) are calculated by the vertical summation of the “Total Net Benefits” column. Total net benefits of the proposed standard are estimated to range from $364.9 million to $385.1 million.

As noted in Table 6 and in previous sections of this analysis on benefits, expected benefits accruing in future years have been discounted to their present value using a 3 percent discount rate to reflect society's time preference. In accordance with OMB guidelines on benefits calculations, calculations have also been made using a 7 percent discount rate. Using this higher rate, total net benefits of the proposed standard are estimated to range from about $260 million to $281 million over the life of complying upholstered furniture produced in a year.

Analyses using both discount rates assume that manufacturers would use FR treatments in a manner that poses no additional risk of injury or adverse health effects to consumers.

Aggregate benefits ranging from about $316 million to $319 million minus aggregate costs ranging from about $34 million to $59 million (midpoint of range).

b. Sensitivity Analysis

The previous discussion compares benefits and costs of the proposed standard using discount rates of 3 percent and 7 percent to express expected benefits accruing in the future in their present value, an estimated value of a statistical life of $5 million, and an estimated average cost of injury of $146,740. Net benefits were also estimated based on estimated increases in costs of producing and marketing furniture that complies with the proposed standard. In addition to these factors, the estimation of benefits was based on assumptions regarding the effectiveness of the standard at reducing losses from cigarette and small open flame ignitions. This section examines the effect of changing any of these assumptions on the expected net benefits that would result from compliance with the proposed standard. In all cases, the estimated net benefits of the proposed standard remain positive.

Discount rates of 3 percent and 7 percent were used to express expected benefits accruing in the future in their present value. Using a 3 percent rate, total estimated benefits of the standard range from about $419 million to $424 million, the range of estimated total costs is about $34 million to $59 million, and total estimated net benefits range from about $365 million to $385 million. Using a 7 percent discount rate, the present value of benefits would range from about $316 million to $319 million, and total net benefits would range from about $260 million to $281 million.

Estimated benefits of the proposed standard were based on a value of a statistical life of $5 million. If benefits are calculated based on a lower bound of $3 million as the value for a statistical life,

total estimated benefits of the standard would range from about $267 million to $270 million using a 3 percent discount rate and about $201 million to $203 million using a 7 percent discount rate. Total estimated net benefits would range from about $211 million to $233 million using a 3 percent discount rate and $144 million to $167 million using a 7 percent discount rate. Alternatively, if a value of $7 million is assigned to a statistical life, the total estimated benefits would range from about $572 million to $578 million (at a 3% discount rate) and about $430 million to $435 million (at a 7% discount rate) and total estimated net benefits would range from about $519 million to $538 million (at a 3% discount rate) and $376 million to $396 million (at a 7% discount rate).

Viscusi, W. Kip.

op. cit.

Estimated benefits of the proposed standard are based on an average societal cost of $146,740 per injury. Changing the estimate used for the cost of injury will have minimal impact on the results, because the share of benefits from reduced injuries is less than 4 percent of total benefits. Hence, even if there were no reduction in injuries from the proposed standard, the total estimated benefits would be about $404 million to $409 million and total net benefits would be $350 million to $370 million using a 3 percent discount rate. Using a 7 percent discount rate, estimated benefits would range from about $305 million to $308 million and estimated net benefits would range from about $249 million to $271 million.

Section 6 of this analysis addresses the expected costs of the standard. Estimates of costs are based on judgments regarding changes to materials that will be required to meet performance tests of the proposed standard, the costs of those changes per unit, and the number of affected furniture items produced annually. Based on the midpoints of ranges of estimated cost impacts of material changes, and different assumptions of reliance on FR fabric treatments as a means to compliance, aggregate costs of the standard were estimated to be $34 million to $59 million for annual production of upholstered household furniture. With these costs, total estimated net benefits of the proposed standard range from about $365 million to $385 million using a 3 percent discount rate and $260 million to $281 million using a 7 percent discount rate. Even if we assume that the costs of the standard are twice those estimated in Section 6 (i.e., $68 million to $118 million) the standard would still have estimated net benefits ranging from about $306 million to $351 million from annual production of upholstered furniture if future benefits are discounted at 3 percent, and about $190 million to $237 million if a 7 percent discount rate is used.

Estimated benefits of the proposed standard were based on assumptions regarding the effectiveness at reducing societal costs of cigarette and small open flame ignitions of furniture. However, if we assume that the standard will have one-half the effectiveness that our estimated benefits are based upon, aggregate benefits would still range from about $210 million to $212 million, and net benefits would range from about $153 million to $176 million, using a 3 percent discount rate. Using a 7 percent discount rate, estimated benefits would

range from about $158 million to $160 million, and net benefits would range from about $100 million to $124 million.

c. Impact of the Proposed Standard on Retail Prices

The estimated costs of the proposed standard include the increased costs of materials, labor, and distribution directly attributable to the rule. It is likely that manufacturers will pass on at least some of the costs of complying with the standard to the consumer, in the form of higher retail prices. The actual increase in retail prices will depend on the price elasticity of demand for furniture products (i.e., the responsiveness of quantity demanded to the change in price). If demand is highly price elastic, then manufacturers will experience a relatively large decrease in sales of upholstered furniture products in response to a price increase, and their ability to pass on increased regulatory costs to the consumer is limited. If demand is price inelastic, consumers respond less intensely to price increases, enabling producers to successfully pass through cost increases.

Regarding the market for upholstered furniture, it is anticipated that demand is relatively price elastic in the short run, because consumers can usually postpone the purchase of a durable good. Increases in retail prices are thus likely to be limited. In the long run, demand is less elastic and any attempt to pass through increased costs is more likely to succeed. Consequently, increases in retail prices are more likely to be observed.

In the absence of information on the price elasticity of demand for upholstered furniture products, it is possible to make use of traditional industry markup rates to provide an upper bound estimate for retail price increases. Such estimates may be viewed as upper bound estimates because they do not reflect the price elasticity of demand. Moreover, traditional markups do not factor in the role of competition, which can also influence attempts to increase prices. Rather, the markup simply reflects the price that producers will want to charge based on historical accounting costs. As noted above, an increase in price will result in a reduction in sales and in the case of highly elastic demand, revenues will decline as well, which will tend to moderate attempts to increase retail prices.

According to industry sources, higher production costs for materials and labor could result in retail prices that are higher by a factor of 2.5, or 150 percent. Based on this markup, the average retail price impact of the proposed standard on furniture items made with FR treated fabrics could be about $23 (for perhaps 2 to 10 percent of all items), and the average retail price impact for furniture produced with barrier materials could be about $48 (for perhaps 4% of furniture items). The average retail price impact for furniture that will not be made with FR fabric treatments or barriers under fabrics (perhaps 86 to 94% of units), could be under $1 per unit. The average increase in retail prices for all upholstered furniture is estimated to be less than $5 per item, based on the traditional industry markup rates.

8. Alternatives to the Proposed Standard

a. The Staff's 2005 Draft Standard

The aggregate benefits of the staff's 2005 draft standard (i.e., the reduction in the societal costs associated with complying furniture), based on the annual sales of a little over 30 million furniture items, are expected to be about $597 million. Total aggregate costs of the 2005 draft standard for each year's production are estimated to range from about $167 million to $184 million, with a midpoint of about $176 million. Although the 2005 draft standard would be expected to increase the use of FR chemicals in the production of urethane foam cushioning and fabrics to achieve compliance, estimates assumed that these chemicals would be selected and used in a way that would not lead to appreciable societal costs. If the use of these chemicals would have adverse health or environmental impacts, the costs of the 2005 draft standard are understated. Estimated benefits and costs per unit would vary greatly depending on cover materials. Most units would incur costs related to FR-treatment of filling materials, and an estimated 10 percent of units covered with more ignition-prone fabrics would require modifications (FR-fabric treatment or FR barriers) that would lead to higher costs of compliance. Projected annual net benefits to society from the staff's 2005 draft standard total $421.5 million. A sensitivity analysis of several factors (value of life, injury costs, effectiveness, and costs) showed that alternative assumptions still yield substantially positive net benefits.

b. The Draft Small Open Flame Ignition Standard

As an alternative to the proposed standard, the Commission could adopt the standard drafted by CPSC staff in 2001 that focused on small open flame ignition of upholstered furniture. That draft standard was the subject of a staff briefing package submitted to the Commission in October 2001. Compliance with the draft small open flame standard would require the use of upholstery cover materials that do not sustain combustion following exposure to a small flame for 20 seconds, or, alternatively, the use of materials that would pass an open flame barrier test. The staff estimated that most fabrics would fail the 20-second flame test unless they would be treated with FR chemicals. Although the FR treatments under that standard specifically addressed small open flame ignition hazards, CPSC staff testing data also showed substantial improvement in cigarette ignition resistance. In fact, most of the estimated benefits of the small open flame standard were projected to result from reductions in societal losses from cigarette ignitions.

Based on estimated costs of compliance and estimated reductions in both small open flame and cigarette ignition hazards, adoption of the 2001 draft small open flame standard would result in estimated aggregate benefits totaling $651 million and aggregate costs of about $272 million from annual production of about 30.5 million pieces of upholstered furniture.

Therefore, estimated aggregate net benefits of the small open flame standard would be $379 million. This compares with estimated net benefits of $365 million to $385 million for the proposed standard.

Smith, Charles, 2001

op. cit.

Based on “Best Estimates” of reductions in ignition propensity and midpoints of estimated increases in manufacturing costs; as with the current analysis, distribution costs are estimated to be an additional 10 percent. The best estimate for cigarette ignition reduction involving cellulosic fabrics is 75%, based on 2003 estimates made by Mark Levenson, EPHA, CPSC.

The net benefits of the staff's 2007 draft standard may also be un

This text is long and has been trimmed here. Open the source document for the complete record.

This is a copy of a public record, reproduced as it was published. It is not legal advice, and it may not be the version a court would rely on. Check the official source before you cite it.