The Habitat for Humanity ReStore system: sourcing and sales of donated wood-based building...

Abstract

The Habitat for Humanity (HFH) ReStore system is composed of hundreds of retail stores that sell a wide variety of building materials. New and used building materials are donated to ReStores and are then resold to generate revenues to support Habitat's mission of building affordable housing for low-income families. This project examined the sales, marketing, and sourcing of wood-based building materials sold through Habitat's ReStore system. Data were collected through the use of a mail survey to all known ReStores currently selling building materials within the United States. Results indicated that on average, wood-based materials made up nearly 30 percent of 2004 sales. The wood-based product category cited as most difficult to sell was window frames; the greatest challenge to selling wood-based materials was missing pieces or components. Several marketing-related variables, such as having a written marketing plan and tracking store performance, were positively related to store sales. Implications are drawn for managers of used building materials enterprises seeking to improve store performance.

**********

Although much focus has been given to curbside recycling during the past two decades, a new emphasis on the reuse, recovery, and recycling of building materials has recently emerged. With building materials currently contributing a high percentage of materials sent to landfills (Neff 2004), the reuse of construction debris and recycled building materials is gaining increased attention (e.g., Falk 1997, 2002; Janowiak et al. 2005). The pure volume of building-related waste is also of concern; an estimated 136 million tons of building-related construction and demolition (C & D) waste was generated in the United States in 1996, with approximately 48 percent coming from demolitions and 44 percent from renovations (Franklin Associates 1998).

The environmentally friendly options of reuse and recycling of these materials help in conserving natural resources and decrease the need for landfills and other disposal facilities. As saving resources by diverting reusable building materials from premature disposal increases in popularity, more retailers are emerging that tailor their operations specifically to this market. Also, organizations have been formed, such as the Building Materials Reuse Association (BMRA), to help educate these retailers as well as the general public about the acquisition and redistribution of used building materials. One charitable organization that deals extensively with used wood-based building materials is the Habitat for Humanity (HFH) ReStore system.

This study used data from the HFH ReStore system to better understand the demand for and marketing of wood-based building materials, and to gain insight into the stores' promotional efforts and management. Findings have implications not just for Habitat for Humanity, but also for the used building materials retail industry as a whole.

Habitat for Humanity ReStores

Perhaps the most widespread, fastest growing, and best-recognized retail "chain" dealing with used wood-based building materials is the Habitat for Humanity (HFH) ReStore system. The HFH ReStore system was established as a fundraising medium where building materials (both new and used) are donated and then resold to the public at discounted prices, with the profits going toward local HFH Affiliate homebuilding efforts. Thus, not only is the ReStore system operated to support the overall mission of HFH to provide affordable shelter for low-income families, it is also designed to support resource efficiency as an avenue to "reuse and recycle" used building materials.

Many HFH affiliates effectively operate ReStores throughout the United States and Canada. In fact, some claim to raise enough funds to build more than 10 additional houses a year because of ReStore revenues (www.Habitat.org). Materials found in the ReStores are often supplied by demolition crews, contractors, building supply stores, and individuals. A large percentage of the used building materials inventory is wood-based, with items ranging from those with obvious uses such as hardwood flooring to less specific uses such as old timbers. The relative value of items in inventory at a given time may range from very high value, such as antique fireplace mantels, to low value items such as scrap framing lumber.

Because the ReStores only accept donations, their inventory is constantly changing. This makes it nearly impossible to predict what items may be available in the future and thus demonstrates that the ReStores. like other used building materials retailers, cannot be run in the same manner as traditional retail stores where stock keeping units (SKUs) and inventories are carefully managed and where quantities can be estimated and ordered well in advance.

Research methodology

This research was survey driven and followed data collection methods as outlined in Dillman (1978) and Fowler (1993). A questionnaire was mailed to store managers at all known HFH ReStores across the United States during Spring 2005. The list of ReStores was obtained from HFH headquarters. The initial mailing of surveys was followed approximately 1 week later by reminder postcards. One week after that all remaining nonrespondents were sent another questionnaire. Other follow-up methods such as phone calls were also made to non-respondents.

Response rates and bias

Of the 268 surveys sent out, 120 were completed and returned during a 2-month period. The total population of ReStores was subsequently adjusted to 215 due primarily to flaws in the address list that was provided by HFH. For example, there were at least 66 wrong addresses, discovered by the return of the initial survey through the mail, or through phone calls. In addition, 10 stores did not sell building materials, thus reducing the actual number of relevant HFH ReStores in our sample frame to 205. Therefore, the adjusted response rate was 53.6 percent (120 out of 205).

Tests for non-response bias were conducted by comparing the first 25 percent of respondents with the last 25 percent by first using a Chi-square test (0.05 level of significance) for the following variables: whether they sell building materials and whether they do deconstruction. Tests showed no significant difference between the groups of respondents. Also, analysis of variance (ANOVA) was used to compare early vs. late respondents on the following variables: sales volume, store size, years of retail experience, and authority over daily operations. Because of the results of these tests, concerns about non-response bias were reduced (Fillion 1975, Armstrong and Overton 1977).

Results

Respondent profile

ReStore managers who responded to the survey had worked in ReStore management for an average of 2.67 years. However, they averaged 11.3 years of retail experience, and thus have considerable experience in retail settings. Managers also rated themselves on how much authority they have over the day-to-day operations of the ReStore. The average rating for this was 8.68 on a scale from 1 to 10, with 1 = none at all and 10 = absolute authority. Such a rating suggests a relatively high level of self-perceived authority pertaining to daily operations.

ReStore profile

A profile of the ReStores indicates that they have more full-time employees (2.24) than part-time (1.69), although both numbers seem relatively low. However, there is a considerable amount of volunteer help (an average of 290 volunteer hours per month). The ReStores vary widely in terms of square footage, with the smallest being a mere 500 [ft.sup.2] and the largest being 39,000 [ft.sup.2]. The average store size is slightly greater than 10,300 [ft.sup.2].

ReStore mean sales volume in 2004 was $187,388; the smallest HFH ReStore sold only $3,960 worth of merchandise while the largest sold $780,000 that year (Table 1). This results in an extrapolated sales volume of approximately $38.41 million for all 205 ReStores in 2004. By contrast, the average "small-market" Lowe's store is approximately 94,000 [ft.sup.2] and has annual sales of $20 million to $25 million.

Although the average store age is 5 years, some are a year or less old (n = 12), with the oldest being open for 18 years (Table 1). More than half of the ReStores have operated for less than 5 years (57.6%), while 28.8 percent have operated for more than 5 but less than 10 years, and only 13.6 percent have operated for more than 10 years. This may be a factor explaining why the average manager-tenure in the ReStore system is low. Table 2 also shows that most ReStores do not devote a high percentage of gross sales to advertising and promotions, with the mean being only 2.89 percent.

All materials sold at ReStores are donated (i.e., they don't purchase any materials in inventory), and we sought to understand the sources of wood-based materials given to the stores. The segment accounting for the largest portion of wood-based items donated to the stores was individual households, with nearly 50 percent of all wood-based items received by the ReStore system provided by this group. Construction firms/builders (18.5%) and large retailers (e.g., Lowe's) (16.1%) also donated significant volumes of materials. An important source of higher value items that produce significant revenue is the original manufacturer (10.3%). For example, makers of hardwood floors (e.g., Bruce Hardwood Floors) and cabinets often donate truckload quantities of new materials that provide important sources of revenues for the stores. Such materials may be slightly damaged or are discontinued styles. Not surprisingly, managers noted that the wood-based products for which demand most greatly exceeded supply were cabinets, hardwood flooring, and 4 by 8 paneling (e.g., plywood and oriented strandboard).

Unfortunately, all ReStores must contend with items that they do not wish to have donated. Survey results indicated that scrap lumber, especially with nails or any type of damage/defects, is a type of donation that people often want to give but that most stores would rather not receive (and many will refuse to accept). Managers also noted that the supply of older windows (especially single pane) and hollow-core, interior doors greatly exceeded demand for these items.

ReStore sales and marketing

The mean sales per square foot for all ReStores is $26.39, or $28.00 for stores open more than 1 year. These figures are low compared to averages for typical building materials dealers. For example, at "big-box" stores such as a Home Depot or Lowe's, average sales can be over $300/[ft.sup.2], with their smaller stores reaching $200/[ft.sup.2] (First Research 2005). It is noteworthy, however, than one ReStore did report a sales per square foot figure of more than $333.

We segmented ReStores by sales per square foot and divided them into three equal sets. The top third were labeled as high performers (high sales per square foot), the middle third were average performers, and the bottom third low performers (low sales per square foot). The differences between these three categories are great, with the average sales per square foot for the high performers at 54.63, compared to 16.81 for average performers, and 8.26 for the low performers.

ANOVA tests were used to evaluate the relationships of sales per square foot and other variables, controlling for ReStore age (Table 2). Variables found to have a significant positive effect on the stores' sales per square foot when controlling for age include: community awareness, the tracking of marketing performance, knowledge to effectively market the store, and having a written marketing plan. All of these are significant at the 0.05 level.

Sales of wood-based materials

We also sought to estimate the revenue attributable to wood-based materials at the ReStores. To do this, we looked at the percentage of total 2004 sales that wood-based materials accounted for and calculated the mean percentage for all stores. Out of 98 valid responses, the mean was 29.19 percent for the percentage of 2004 sales that were attributable to wood-based materials (Table 1).

[FIGURE 1 OMITTED]

ReStore managers were also asked to indicate which wood-based material they considered the most difficult to sell. Respondents were asked to circle one choice from a number of categories of wood-based building materials (Fig. 1). The material chosen with the most frequency was window frames; chosen by 32.3 percent (n = 30) of the respondents, followed by countertops (n = 13). Respondents noted that windows were difficult to sell due to sizing issues and damage. The least difficult to sell categories of wood-based materials appeared to be antique items and framing lumber since they were not selected by any respondents.

Also examined were the speed of sale and relative profitability for these same categories of wood-based materials (Fig. 2). Speed of sale (e.g., inventory turnover) was measured on a scale from 1 to 5 where 1 = slow and 5 = fast. Framing lumber was ranked as fastest for speed of sale (4.29), with the slowest being window frames (2.27). Relative profitability was similarly measured on a 5-point scale, where 1 = low and 5 = high. The wood-based material with the highest ranked relative profitability was cabinets/cabinet parts (4.25), and the lowest was once again window frames (2.56).

The survey also asked about the extent to which six issues hindered managers' ability to sell wood-based materials already in inventory. Managers were asked to rate this extent on a 5-point scale, where 1 = no challenge and 5 = great challenge (Table 3). Results show that the biggest problem (the issue with the highest ranking) was pieces/components missing (3.34), with the lowest being customer unsure of usage (2.47).

Discussion

The high ranking that managers gave themselves (8.68 of a possible 10) in terms of daily authority over store operations was consistent with our exploratory findings during pre-data collection discussions with ReStore managers. Also as expected was the low percentage of gross sales devoted to advertising and promotions (<3 percent of total sales in 2004). Clearly, ReStore managers view themselves as having the authority to make income-producing decisions yet they are not choosing to make serious commitments to advertising and promotions. However, based on earlier conversations with Habitat leadership and store managers, we had expected that the mean percentage of 2004 revenue attributable to wood-based materials would be greater than what we found it to be (29%). The proportion of non-wood-based materials sold in used building materials stores should be considered when devising a marketing strategy for the ReStores, especially since non-wood materials may require different marketing techniques compared to the ways wood-based materials are marketed.

[FIGURE 2 OMITTED]

The category of wood-based materials most difficult for the ReStores to sell is window frames. This product was also ranked as having the slowest speed of sale and lowest relative profitability. Used building material retailers should reconsider their acceptance policy with regard to such undesirable materials, and/or help to find alternative uses for window frames or recycling options. When possible, floor space should be devoted to faster-selling, more highly profitable items such as architectural salvage and cabinets/cabinet parts. However, before deciding to eliminate a category altogether, such as window frames, management must evaluate how it compares to other materials. Although window frames may be difficult to sell as compared to other wood-based materials, we do not know how they compare to the non-wood-based materials outside the scope of this study.

As expected, results suggest that our marketing variables, such as the tracking of marketing performance, knowledge to effectively market the store, and having a written marketing plan, all have a significant effect on ReStore sales per square foot when accounting for store age. However, many managers are not currently engaging in marketing planning and promotional activities that could increase sales. Knowing this, the HFH ReStore system may consider the importance of effective marketing strategies and the necessity of implementing marketing activities at their ReStores.

Limitations

There are certain limitations to consider with this study. First, the mailing list obtained from Habitat for Humanity International had inaccuracies and likely did not include the entire population of ReStores. Moreover, it is estimated that by late 2006 there will be as many as 500 ReStores operating in the United States, which implies a need to conduct follow-up surveys to see how the data change over time. Finally, since the data were obtained through the use of surveys, there are possible biases based on various characteristics of the respondents.

Conclusions

Used building material dealers have a small, but growing place in the market for wood-based building materials. The Habitat for Humanity ReStore system is the largest single player in this evolving industry, and this project has served to increase our understanding of the system and the wood-based materials that it sells. The study shows that ReStore inventory is composed of wood-based building materials with a wide range of profitability and speed of sale. Thus, ReStore managers must learn to effectively manage their building materials inventory to maximize both the societal good such stores provide as well as the revenue potential for the HFH ReStore system.

The ReStore system is generally young, with the majority of ReStores less than 5 years old and many more stores in the planning stages. By learning how to effectively manage and market their materials and their stores, ReStore managers may achieve superior growth in store size, sales volume, and profitability. ReStores are examples of how an organization can attempt to achieve a "triple bottom line" (Willard 2002) result by seeking not just economic success, but also social/ethical, and environmental results.

Literature cited

Armstrong, J. and T. Overton. 1977. Estimating nonresponse bias in mail surveys. J. of Marketing Res. 16:396-402.

Dillman, D. 1978. Mail and Telephone Surveys: The Total Design Method. John Wiley, New York.

Falk, R. 1997. Wood recycling: Opportunities for the woodwaste resource. Forest Prod. J. 47(6): 17-22.

______. 2002. Wood-framed building deconstruction: A source of lumber for construction? Forest Prod. J. 52(3):8-15.

Fillion, F.L. 1975. Estimating bias due to nonresponse in mail surveys. Public Opin. Q. 40:482-492.

First Research. 2005. Industry profiles: Building material supply. www.firstresearch.com/Customers/reports/buildingmaterialsupply.asp. Accessed on 7/12/05.

Fowler, F.J. 1993. Survey Research Methods. Sage Publications, Thousand Oaks, CA.

Franklin Associates. 1998. Characterization of building-related construction and demolition debris in the United States. Rept. No. EPA530-R-98-010 prepared for the U.S. Environmental Protection Agency, Washington, DC.

Janowiak, J.J., R.H. Falk, B. Beakler, R. Lampler, and T. Napier. 2005. Remilling of salvaged wood siding coated with lead-based paint: Part 2. Forest Prod. J. 55(7/8): 81-86.

Neff, T. 2004. Recycled building materials in Boulder, Colo, get EPA award. Knight Ridder Tribune Business News, September 2, p.1.

Willard, B. 2002. The sustainability advantage: Seven business case benefits of a triple bottom line. New Society Press. www.newsociety.com/.203 pp.

Amy R. Gresock

Judd H. Michael*

Ann E. Echols

Paul M. Smith*

The authors are, respectively, Graduate Student, Dept. of Management, Univ. of Central Florida, Orlando, FL (amy.gresock@bus.ucf.edu); Associate Professor, Sustainable Wood-based Enterprises, School of Forest Resources, The Pennsylvania State Univ., University Park, PA (jhm104@psu.edu); President, green WELLS, Inc., State College, PA (A.Echols@greenwells.net); and Professor, Forest Products Marketing, The Pennsylvania State Univ., University Park, PA (pms6@psu.edu). Funding for this research was provided by Penn State's Social Science Research Institute. This paper was received for publication in November 2005. Article No. 10139.

* Forest Products Society Member.

Table 1. -- Profile of Restores.

Item                   n    Min.      Max.        Mean        SD (a)

Full-time employees    113      0          25           2.24        2.87
Part-time employees    102      0           9           1.69        1.83
Volunteer hours per    104      0       2,000         289.89      366.65
  month
Store space            115    500      39,000      10,305.13    8,771.63
  ([ft.sup.2])
Store age (yr)         118      0          18           5.1         3.97
2004 sales volume ($)  107  3,960     780,000     187,387.62  175,432.59
% 2004 sales from       98      0         100          29.19       21.87
  wood-based
  materials
% gross sales for      103      0          51           2.89        5.51
  adv.-promo.
Sales/[ft.sup.2] ($)   103      0.21      333.33       26.39       41.78

(a) SD = standard deviation.

Table 2. -- ANOVA analyses of the relationships of marketing variables
and store age to sales per square foot. (a)

                         Sums of  Degrees of  Mean
Sources of variation     squares  freedom     square  F       p-value

ANOVA for the relationships of store age and community awareness to
sales/[ft.sup.2]
Total                    131817   92          1433
Among all categories      74304   40          1858     1.68   0.039
Age                       26058   16          1629
Comm. awareness            3034    4           759
Age x comm. awareness     43591   20          2180     1.971  0.026
Within categories         57513   52          1106

ANOVA for the relationships of store age and tracking performance of
marketing activities to sales/[ft.sup.2]
Total                    131817   92          1433
Among all categories     126010   52          2423    16.69   0.000
Age                       31380   16          1961
Tracking performance       8652    6          1442
Age x tracking perform.   89678   30          2989    20.59   0.000
Within categories          5807   40           145

ANOVA for the relationships of store age and knowledge to market the
store to sales/[ft.sup.2]
Total                    131786   90          1464
Among all categories     125016   51          2451    14.12   0.000
Age                       28940   16          1809
Knowledge to market        6288    6          1048
Age x knowl. to market    91062   29          3140    18.09   0.000
Within categories          6770   39           174

ANOVA for the relationships of store age and written marketing plan to
sales/[ft.sup.2]
Total                    131789   91          1448
Among all categories     118136   53          2229     6.200  0.000
Age                       28922   16          1808
Written marketing plan    15258    6          2543
Age x written mkt. plan   75226   31          2427     6.750  0.000
Within categories         13653   38           359

(a) p-values are significant at the 0.05 level.

Table 3. -- Degree of challenge posed by each problem in selling wood-
based materials. (a)

Problem                                 n    Average rating

Pieces/components missing               103  3.34
Damage/defects                          104  3.27
Insufficient space in the store         106  3.22
Merchandising limitations within store  104  2.98
Product "out of style"                  104  2.61
Customer unsure of use                  101  2.47

(a) Rated by respondents on a scale from 1 to 5. where 1 = no challenge
and 5 = great challenge.