Environmental Valuation & Cost-Benefit News

Link: http://temp.sfgov.org/sfenvironment/aboutus/innovative/greenbldg/gb_productivity.pdf

One of the first studies to document the correlation between comfort and productivity was produced by researchers at Rensselaer Polytechnic Institute in a study of the West Bend Mutual Insurance Company's
headquarters, completed in 1991. In this new office building, workstations were equipped with comfort control systems (Personal Environments Modules-PEMs), manufactured by Johnson Control Inc. The PEMs
allow employees to control temperature, ventilation, lighting, and sound masking to suit their personal preferences at any given time. The company had monitored employee productivity prior to the move, so researchers were able to compare the productivity for occupants in the new building with that measured in the older, more conventional building. The research showed a median productivity improvement of 16% compared to the old building, and that 2.8% of this improvement could be attributed to benefits of the PEMs themselves.

Several institutions are conducting research on the effects of occupant satisfaction and indoor air quality (IAQ) on productivity. The Center for Indoor Climate and Energy at the Danish Technical University (DTU)
performed laboratory tests to evaluate the effects of IAQ on the performance of simulated office tasks such as typing, proofreading, and addition. Results indicated that decreasing airborne pollutants can improve the performance of these tasks up to 6%. The airborne pollutants were provided by a swatch of 20-year old carpet from a building with a history of sick building syndrome symptoms, replicating conditions that may persist in many office environments.

Increasing evidence shows that people prefer moderate levels of patterned complexity and sensory variability in the environment (Humphrey, 1980; Platt, 1961). An environment devoid of sensory stimulation and variability can lead to boredom and passivity (Cooper, 1968; Schooler, 1984). The sensory variability in the natural environments also enables people to actively select conditions that fit their current needs and desires – e.g., the warmth of the sun when they are cold, or the comfort of a tree canopy if they are too warm. Building interiors seldom offer such personally adaptive possibilities.

Stress Reduction, Health and Well-being Studies by Roger Ulrich (1993, 1984) have consistently found stress reducing and health promoting
outcomes associated with passive viewing of nature stimuli through windows, videos, and photographs. R. Kaplan (1992) reports similar results in a field study of office workers. Kaplan found that workers who had window views of nature felt less frustrated and more patient, and reported more overall life satisfaction and better health than workers who did not have visual access to the outdoors or whose view consisted of built elements only.

In contrast to the enhancing effects of nature, buildings can also have negative impacts on health. Of particular importance are ambient conditions, including noise, temperatures, air quality, and lighting.
Uncontrollable noise or excessive temperature conditions have been linked to stress symptoms and irritability (Heerwagen et al, 1995), while lighting that produces glare or visual discomfort is more likely to
be associated with headaches and eye problems (NRC, 1983). The problem that has received the most attention is indoor air quality. Building attributes that directly influence health include air borne toxins and noxious chemicals associated with materials and finishings, cleaning products, and equipment; noxious and hazardous substances associated with work processes; and air borne pathogens associated with HVAC systems.
Enhanced indoor air quality from reduced use of toxic and noxious materials is a core feature of green buildings. Improved IAQ is also central to the arguments about ancillary benefits. According to a model
developed by Fisk and Rosenfeld (1997), improved air quality could result in significant reductions in illness and absenteeism associated with respiratory disease, asthma and allergies, and sick building syndrome
symptoms. They estimate productivity gains of $17 billion to $164 billion annually associated with improved air quality. They also estimate improved worker performance from enhancements of the thermal
environment and lighting to be in the range of $12 billion to $125 billion annually.

Most of the studies have focused on emotional states and stress reduction (see Ulrich 1993 for a review). However, there is growing evidence of a strong linkage to cognitive functioning. For instance, a recent experiment by Lohr (Lohr et al, 1996) found that subjects working in a windowless room with plants worked more efficiently, had lower blood pressure readings, and felt more attentive than subjects working in the same room without plants. In another study varying window views, Tennessen and Cimprich (1995) found that people whose view was predominantly natural (as opposed to built) scored better
on tests of directed attention. Hartig and colleagues (1991) report similar results in a field experiment. People who went for a walk in a predominantly natural setting performed better on several attentional tasks than those who walked in a predominantly built setting or who quietly read a magazine indoors. These results, coupled with work on stress reduction by Ulrich (1993) strongly indicate that green buildings that incorporate natural vegetation are likely to have positive benefits on both well being and performance. At the present time, the exact psychological and physiological mechanisms that lead to such effects are not known. However, Ulrich suggests that a critical dimension is the relationship between contact with nature and positively toned moods.
Positive moods turn out to be critical for a wide range of outcomes of interest to this study, including job satisfaction, work involvement, motivation, organizational attachment and lowered absenteeism (Mitchell,
1989). Clark and Watson (1988) found that positive moods are associated with the physical setting at work and with daily events (especially social interactions among workers), while negative moods are associated
with daily hassles (such as discomfort and distractions).

Studies by Alice Isen (1990; Isen et al, 1987; Moore and Isen, 1990) show that positive moods also have numerous beneficial impacts on cognitive and social functioning. Isen’s studies have found that positive
moods are associated with:

• Enhanced discriminative learning and more efficient decisional processes on complex tasks.
• Greater use of inductive rather than analytic heuristics in problem solving.
• Higher joint benefits in negotiations and more innovative approaches.

Complex cognitive strategies are less likely when people are depressed, unhappy, or stressed. This is because negative moods or stress tend to restrict attention and lead to stereotypic responses. Daily irritants,
including discomforts, tend to generate high negative affect and thus are likely to interfere with higher level cognitive functioning (Clark and Watson, 1988). Isen (1990) speculates that positive moods increase the tendency to “break set” and to see relatedness between divergent events or appearances. This is because feeling good promotes diffuse rather than focused attention and this leads people to see things differently (e.g. people notice more details) or to search more broadly for solutions and alternative interpretations. Joseph LeDoux (1996), one of the nation’s leading brain researchers, cites neurological evidence to support this hypothesis. He has found that positive feelings lead to heightened activity of the right parietal brain region – the section of the brain that is associated with a more global, expansive cognitive style. Thus, positive feelings directly affect brain processes related to
performance on tasks requiring creativity and novel problem solving.
Building Features and Cognitive Functioning.

Although Isen and LeDoux do not discuss the specific aspects of the environment that influence positive feelings and cognitive processes, we know from the research on nature that non-threatening, aesthetically
pleasing natural settings elicit positively toned moods and physiological restoration. It is not unreasonable to expect that similar results would be true of buildings that incorporate naturalistic features, either literally (through contact with nature) or through design features that mimic the key aesthetic and sensory characteristics of preferred natural environments (Orians and Heerwagen, 1992; Hildebrand, 1991).

Another potential mechanism linking building attributes to performance is cortical arousal. For tasks requiring high levels of cognitive effort, the ability to access sources of “cognitive tranquillity” may be an effective way to keep arousal at a desired level, if it begins to exceed levels conducive to the task (Coss, Clearwater, et al, 1989).

Environmental features that support cognitive tranquillity include perceptual distance and expansiveness. This can be real (as in a distant view of the landscape through a window), or it can be virtual (as provided by the design of vertical surfaces in the workspace surround). Coss and Clearwater found that distant views and view corridors with interesting focal points are even especially powerful in promoting cognitive rest. Focal interest can be created as part of a green design strategy, particularly through lighting design and windows.

Personal control has assumed such a powerful place in environmental psychology that few question what it means or what people really want to control. The research on the advanced comfort workstation produced by Johnson Controls supports this notion (Kroner et al, 1992). The research tracked workers in an insurance company as they moved to a new building with advanced controls workstations. The study found that productivity
increases of 2.8 % could be attributed to the new workstation. Interestingly, the total productivity increases associated with the move to the new building were substantially higher (16%). Although the research report does not try to link the overall productivity increase to the building features, it is evident from the pictures and from the report descriptions that the new building had a number of amenities that were lacking in the old building. Especially relevant is the presence of an attractive landscape with a pond, extensive windows and daylight, and a more open and spacious interior. Another key difference is the location of the workstations: in the new building, 96% of the workers with the personalized control workstations were located on the perimeter with window views. In contrast, only 30% of the workers in the old building had access to windows. The remainder were in interior core spaces.

As the economy continues to grow and unemployment decreases, many U.S. companies are discovering that finding and retaining high quality employees is not as easy as it used to be. Turnover is costly to any company (Phillips, 1990), but especially in knowledge fields where the “product” is human brainpower that goes with the worker when he/she leaves. In this new climate, building design that contributes positively to human well being and performance may be perceived by decision-makers as one of many strategies to attract and retain
workers.

There is growing evidence that buildings are used strategically as a sales and marketing tool (Petzinger, 1997) and as an employee “benefit” to attract and retain high quality workers (Becker and Lynn, 1986). In addition, the building itself as a symbol of the corporation’s environmental and social performance may be a powerful attraction for potential employees (see studies reviewed in Turban and Greening, 1997).

Companies also care about absenteeism and health care costs. In a study of 210 salespeople, George (1989) found that people who tend to experience positive moods at work were less likely to be absent.
Further, there are anecdotal accounts of a relationship between improved indoor air, improved lighting, and reduced absenteeism in green buildings (Browning and Romm, 1995). Sustainable practices have gained increasing attention in the mainstream organizational management literature, including the Harvard Business Review (Eccles, 1991; Magretta, 1997) and the Academy of Management Review (Russo and Fouts, 1997). These studies suggest that sustainable design and operations (especially energy efficiency and pollution prevention) can have far reaching impacts on an organization, including:

• Reduced legal and insurance costs associated with reduced risks to current and future generations
• Reduced regulatory inspection load
• Enhanced community livability
• Enhanced relationships with stakeholders
• Process innovation associated with the quest for resource efficiency
• Improved ability to market to pro-environmental consumers

In addition, there is a growing recognition that “green” buildings may play a significant role in promoting the organization as a whole. As noted by Hodgkinson (1993), in an extensive evaluation of companies in Great Britain and Europe: Businesses will increasingly want their flagship premises to present a shining example of environmental
friendliness in terms of energy efficiency, the use of building materials, and the impact on the wider environment….New aesthetics will undoubtedly be developed to make more visible the fact that green
principles have been adopted. Key determinants of these aesthetics may include the use of more durable or recycled materials, or showing off energy efficient plant and passive solar design features, for
example. (pg. 103).

If these hypothesized links between building attributes and features are real, then studies of buildings that vary along critical dimensions should show differential outcomes. This section provides an overview of key results from the DOE funded study of the ancillary benefits of green buildings.

Study Site
The study used a “before” and “after” approach in which data were gathered in both a standard building and a new, green building. Miller SQA in Holland, Michigan, was the case study used in the development of the protocol. Miller SQA, a wholly owned subsidiary of Herman Miller, Inc., is a fast-growing re-manufacturer, manufacturer and vendor of office furniture. It has a niche market providing “just in time” furniture products for small businesses and nonprofit institutions
The 290,000 sq. feet SQA building, designed by William McDonough, is almost half again as big as the old building. It is a manufacturing plant, warehouse, and headquarters building housing approximately 700
people (600 in the manufacturing plant across three shifts, and 100 in the office). The manufacturing plant has about 30-40% temporary workers. Energy efficient aspects of the building include large-scale use of
daylighting, energy efficient fluorescent lamps, daylight controls, energy efficient glazing, and motion sensors in rooms that are intermittently occupied. The HVAC system features state of the art digital controls, including sensors, controllers, and data loggers. Green components include environmentally sensitive materials throughout the building, minimally invasive site utilization (including a wetlands and use of natural field vegetation rather than planted and mowed grasses), enhanced indoor air quality, and extensive recycling. In addition, building materials were obtained locally whenever possible to reduce transportation costs and energy. The SQA building also has a lunchroom, rest areas at each end of the manufacturing area, and a fitness center, including a full size basketball court. Prior to its move into the new green facility, Miller SQA was housed in a smaller industrial type building about five miles away. The “old” building was also divided into an office area and a manufacturing plant. The building had high ribbon windows lining the perimeter walls in both the manufacturing and office areas. There was limited daylight in the building. The building had an employee lounge, a small outdoor seating area with picnic tables, and conference rooms. It did not have an exercise facility.

A total of 262 workers completed the survey in the first building, about 46% of workforce. There were 106 males and 77 females in the manufacturing area, and 32 males and 49 females in the office area. The figures for the new building were 326 workers, about half of the total workforce. This included 138 males and 87 females in the manufacturing area, and 37 males and 38 females in the office area. The manufacturing plant has three shifts: 7 am to 3 PM (the daytime shift); 3 PM to 11 PM (the second shift); and 11 PM to7 am (the night shift).

Individual Level Measures. The data on individual level outcomes were obtained from a questionnaire completed by workers in the old building in October 1995, and in the new building, nine months after the move (in July 1996.) The survey instrument was organized around seven dimensions. The dimensions dealt with a range of satisfaction and perception questions as well as work and environmental experiences.

Workers were asked to rate their comfort and satisfaction with a wide range of ambient, aesthetic, social and functional features of the environment as well as their behavioral, physical, social, and psychological experiences. These included illness symptoms (sore eyes, headache, sore back), psychological well being (fatigue, being in good spirits at work, opportunities for rest and relaxation), job satisfaction, and social relationships. In addition, workers were asked to “grade” the conditions and features of the environment (the
Environmental Report Card). Organizational-level outcomes. In addition to the survey, we tracked a number of organizational outcomes,
using data collected by Miller SQA as part of its Total Quality Metrics Program (TQM).

The TQM dimensions include:
• On time shipment
• Efficiency (percentage of waste or “scrap” from the production process)
• Quality (the ratio of reported defects compared with the number of products shipped)
• Profitability (employees share of net profits in the form of bonuses)
• Productivity (ratio of total labor costs to the dollar value of sales)

The data were collected on a monthly basis. Due to the increases in the number of workers following the move to the new building, the TQM data were normalized for analysis. We did not track turnover or absenteeism. The turnover rate is so low at Miller SQA that there would be insufficient data for analysis purposes. Although we had intended to track absenteeism, the data were not collected in a way that would separate out illness from other reasons for absence (including having a sick child, being on maternity leave, or taking a day off for personal reasons). In order to track absenteeism, we would have had to reorganize the database. This proved to be an insurmountable problem.
Summary of "Before" and "After" Results The data presented in this article are summaries of more detailed reports. For further information see Heerwagen et al, 1997a, b; Heerwagen et al, 1996; Heerwagen and Wise, 1998).

Quality of Work Life. The results from the preliminary survey data analysis indicate that perceptions of and experiences in the two different buildings differed on many dimensions (see Heerwagen et al., 1996; Heerwagen and Wise, 1998). The data show that the new, green building was associated overall with higher quality of work life than the old building, and that it had a more positive impact on perceived work
performance and job satisfaction. The new building was also perceived as “healthier” than the old building: however, there were few differences in self-reported illness symptoms (such as headaches and fatigue). The
exception was a decrease in headaches for office workers in the new building. Whereas 16% said they had headaches often or always in the old building, only 7% did in the new building. There were no differences in
the health and data reported for the manufacturing workers.

The windows and lighting received very high ratings by workers, particularly the presence of extensive interior and exterior views and the high daylight levels. Air quality in the new building was also rated more favorably, although there continued to be problems in the manufacturing area with low air movement in some locations. The environmental features that were rated most negatively were temperature conditions and noise. The negative ratings for temperature conditions, especially complaints of overheating in the summer, came primarily from the manufacturing area, which is not air-conditioned. Noise was a problem in both the plant and office areas, largely due to a couple of high noise manufacturing processes and the openness of the design plan. (Since the survey findings, the high noise tasks were moved to a more distant area of the plant.

In addition, six large fans, three intake and three exhaust, were added to the roof monitors to exhaust the hot interior air and bring in cooler outdoor air.) One of the more interesting findings from the data is the substantial variation in response to the building between the office and manufacturing workers as well as variation among the three manufacturing shifts. These differences are discussed below.

Differences among Manufacturing Shifts
The differences between the manufacturing and office workers, as well as differences among the manufacturing shifts, are likely to be due to multiple factors. However, the features and attributes of the
physical environment also vary in ways that are likely to affect a number of the study outcomes.

The naturalistic amenities associated with nature, windows, and daylight are lacking for those who work at night because they cannot see outdoors and there is no daylight available. Thus, the perceptual differences
between the new and old building environments are not nearly as great as they are for the daytime workers. As a result, the benefits that would normally be associated with these features may not accrue equally among
manufacturing workers. If this is true, we would expect to find differences in outcomes that are particularly relevant to these features, particularly those mediated through psychological mechanisms such as mood at work, work attitudes, and feelings about the work environment.
This is, in fact, what we found. The workers in the night shift had more negative outcomes on all of these dimensions; in fact, their responses in the new building were even more negative than they were in the old
building (Heerwagen, 1998). Daytime workers had the most positive outcomes, and those in the second shift (3 p.m. to 11 p.m.) tended to have outcomes intermediate to the other shifts. (For more details see
Heerwagen and Wise, 1998.)

Differences between Manufacturing and Office Workers

Whereas the differences among the environment experienced by the three manufacturing shifts was largely due to diurnal changes in daylight, the differences in the environments of the office and manufacturing areas
relate to a number of design features. The office and manufacturing areas vary in spatial features, sensory variability, daylighting, windows, visual access, decor, and color. In part, this is due to the actual design, and in part to the way the space is altered by the addition of furnishings and equipment. For instance, high shelves and equipment in the manufacturing area tend to block views to the outdoors and reduce daylight levels in many parts of the space. Large bay doors along the perimeter walls for loading and unloading trucks are kept open all day long when weather permits, even though workers are urged by facilities managers to close the doors so that the air handling system will function efficiently. The workers may be reluctant to close the doors if they provide an important source of contact with the outdoor environment (especially for fresh air, views, and increased daylight).

The office space, in contrast, is air conditioned, carpeted, brightly and evenly lighted, with extensive window views to the outdoors along the entire southwest-facing wall. It is also separated from the noise of the
manufacturing area and is thus quieter. The office spaces also have interior windows looking out on the daylighted “street.” The street is lined with bamboo plants along one side. As a result, office workers can
see greenery on the exterior as well as interior of the building. Many of the offices also have additional plantings and colorful decor. The physical attributes and properties of the office and manufacturing areas are likely to have affected differences in workers’ responses to and experiences of the environment. Comparisons of the manufacturing
and office workers show that office workers had more positive outcomes on almost all dimensions. Office workers rated the aesthetic and naturalistic features of the environment (daylight, windows, contact with
nature, contact with wildlife) more favorably than did the manufacturing workers suggesting greater perceptual experience (see Heerwagen et al, 1996, 1997).

Office workers also had more positive work attitudes, and were more likely to be in good spirits at work. This outcome was true for both buildings. However, both manufacturing and office workers reported
increased fatigue in the new building and increased feelings of being “overworked.” This is very likely due to the large increase in sales associated with increased manufacturing capability.

Although there are likely to be many factors that influence these differences (including the work itself) it is also likely that differences in the physical environments also contributed to the variation in responses (see Heerwagen et al, 1996 for details).
Organizational level measures. Data from the TQM performance measures show high variability in both buildings across all organizational performance measures. This is due to the seasonal changes in sales that
are typical throughout the furniture industry. For four of the TQM measures (effectiveness, efficiency, quality, and total productivity), there were increases in performance for at least five of the eight matched months. There were also modest increases of 0.5 to 2% on these four measures. Although they were not statistically significant, they may be very organizationally significant: any improvement at all can be highly relevant to an organization in a competitive market.

Thus, the data need to be interpreted not only through statistical processes, but also from an organizational perspective. If viewed in this way, the performance results actually show an increase across a number of
dimensions when the data are matched according to the time of year (e.g., comparing December 1995 with December 1996

Can the Results be Attributed to the Building Features?

It is not unreasonable to ask whether the positive results are due to the particular features and attributes of the building highlighted in this article, or to other factors, such as a general “halo” effect associated with being in a new building. Another potential confounding factor is features that are not directly related to green design (such as color and aesthetics, spatial arrangements, new furniture), which could be a big contributor to the workers’ overall responses.

First, if there were a halo effect, we would expect to see it across all variables. This did not happen. The workers’ perceptions and experiences were not all consistently positive. Second, the results from the
manufacturing workers are especially telling. For numerous outcomes, the non-daytime workers actually showed a more negative response to the new building compared to the old one. Another potential explanation could be changes in organizational policies that took place between the old
and new buildings. The biggest difference was the decrease in employees’ share of profits. The profits decreased in the new building because of the costs associated with the building itself. The employees clearly
felt the economic impact of the investment. However, the decrease in profits affected everyone’s paycheck.

Thus, if this were a critical factor, we would expect to a general negative effect. Further, we would expect to find the negative impact widespread and distributed equally across the workforce. This was not the case. And finally, the more negative responses of the non-daytime manufacturing workers could be due to factors associated with working outside the normal daytime hours. This could explain the differences between shift workers to some extent, but it would not explain why the non-daytime workers tended to react more negatively to the new building than to the old one. It is possible that the night and evening workers see the new building as a cost, while daytime workers view it as an investment. Without further research, we cannot determine whether or not this is a reasonable explanation.

The strongest support for our interpretation of the results (e.g., that outcomes are related to the perception and experience of naturalistic features) is the consistent and logical connection between the results of this research and studies conducted by others. We know from the growing body of literature that contact with nature is beneficial to people, even when the contact is second hand through window views. The results of this research suggest that it is not only what is seen out the windows that is important, but what comes in – e.g., daylight, fresh air, nature sounds, and a sense of relatedness to the outdoors.

However, we cannot rule out the possibility that other, non-green factors are influencing workers’ responses. For instance, the aesthetic and spatial features of the building clearly differ between the two buildings and may be having substantial effects. Future research in other buildings, using the before and after approach, is clearly needed.

Conclusions
The conceptual framework presented here suggests that buildings have potentially far reaching impacts on human well being and on organizational effectiveness. More research is certainly needed to verify and extend these connections.

by Mark Palmer and Alicia Mariscal
City of San Francisco SF Environment www.sfenvironment.com
11 Grove Street
San Francisco, CA 94102
415-554-6397

http://temp.sfgov.org/sfenvironment/aboutus/innovative/greenbldg/gb_productivity.pdf