The Hundred-Year Chair · a chapter

Worth Its Footprint

In 2007, BusinessWeek published a critical article about Terrapass, the company I had founded with a group of MBA students three years earlier. Our CEO, my former student Tom Arnold, got a call from his mother asking whether he was okay. Terrapass had started with an argument about the economics of reducing greenhouse gases. Now Tom's mother was worried about him.

Most of our customers were concerned about the environmental impact of driving their cars. Terrapass sold them credits to compensate for their emissions by paying for reductions elsewhere. We aimed to put the money where a reduction cost the least: capturing methane from landfills and farms, for example, or supporting wind power. Someone facing the inconvenience of giving up driving or the expense of buying a hybrid could help pay for a cheaper reduction somewhere else.

The troublesome word was additionality. Had our customers' money caused reductions that would not otherwise have occurred? Perhaps that dairy farmer would have captured the methane from decomposing cow manure anyway. The project could be doing something environmentally useful without our customers' money causing any extra reduction.

I believed in the long-run effect of making emissions reductions more rewarding. A source of revenue could make more projects worth undertaking, even when it was difficult to point to the project caused by one customer's payment. Think about air travel. If I cancel a cross-country trip at the last minute, the plane generally leaves anyway. It may leave with an empty seat. When people consistently buy fewer tickets, airlines eventually respond to the lower demand by offering fewer flights or using smaller planes.

That was hard to communicate. An empty seat is visible. A change in the number of flights an airline eventually offers is much harder to connect to one decision. I did not blame the BusinessWeek reporter for asking what our money had accomplished. Believing that incentives change a market does not by itself establish how many tons of reductions a particular credit represents. Our customers were entitled to ask.

Years later, I was pleased with myself for making chairs from scrap. The old additionality question was still there. What happened differently because I used those offcuts?

If the wood would otherwise have been discarded and the chair replaced one I would have bought, there was a plausible benefit to investigate. If I made an extra chair that nobody used, the scrap had not displaced much. I had also added fasteners, finish, and the work of making it. The zero beside material in my costing notebook was a price, not an environmental assessment.

The old oak I had ordered came from dismantled log cabins. I intended to use it for an even more ambitious chair, one I hoped could last a thousand years. I wanted its history and the chance to work with it. But reclaimed wood still had to be transported and prepared. I did not yet know how much of those hewn logs would become chair parts. Its age could not tell me whether it was a better environmental choice than new wood obtained nearby. That comparison would need the shipping, processing, usable yield, and some account of what would otherwise have happened to the timber.

There was also the scale of my attention. About twenty years earlier, I had calculated the environmental impact of one passenger's round trip by airplane from New York to San Francisco and compared it with disposable foam coffee cups. Under the assumptions I used, the flight came out at about 250,000 cups. I have shown students that calculation ever since. It combines several kinds of environmental damage, not just carbon emissions. Even as a rough comparison, it is hard to reconcile with the attention I give a cup in the trash or a useful scrap of plywood.

But canceling one ticket would not instantly prevent a quarter-million cups' worth of damage. I had assigned a passenger a share of the flight's impact. What would change if that passenger stayed home was the harder question we had faced at Terrapass. The calculation gave me a sense of scale; it did not tell me how quickly an airline would respond to fewer bookings.

None of this made the cup or the scrap irrelevant. It did mean I could spend an afternoon finding a use for a few offcuts while overlooking a decision with much larger consequences. I needed to compare the impacts, not the hours I had spent thinking about them.

For the chair itself, I wanted to get beyond the appearance of the material. The inexpensive plastic monobloc in my costing exercise looked like an obvious candidate to improve on. I compared it with Thonet's bentwood No. 14, using a model built with the machine.

The first pass gave the bentwood a two-to-one win. But when I replaced my estimates of the shipping-container space required per chair with the reported export packing volumes, the margin shrank to almost nothing. The monobloc's compact stack allowed more chairs into each container, spreading the impact of the journey over more seats.

For both calculations I used Eco-indicator 99, an older life-cycle assessment method that comes, pleasingly, from the Netherlands, the country of Rietveld and Eek. I teach it because it is transparent enough to work through by hand. It follows materials and processes through production, delivery, use, and disposal, then combines estimated damage to human health, ecosystems, and resources into a score in ecopoints. The weightings involve judgments; an ecopoint is not a kilogram of carbon dioxide. Here the two chairs arrived at nearly the same total.

Modeled impact from material production through disposal, assuming manufacture in China and delivery to the United States. Pt means ecopoints; 1,000 millipoints (mPt) equal one point. The figures are medians from 200,000 model runs that vary uncertain inputs. Shares are rounded; separately calculated medians need not add to 100 percent. The five- and forty-year service lives are assumptions, not measured results.

The last two rows changed the comparison entirely.

How much impact buys a year of sitting? At the assumed service lives of five and forty years, the No. 14 has about one eighth of the modeled impact per year. The advantage comes almost entirely from the years, not from making the wooden chair more cleanly. Give both chairs the same service life and the near tie returns. I had assumed the very thing I most wanted to achieve.

Still, the arithmetic offered a direction. Hold the initial manufacturing impact constant and extend useful service from five years to fifty: its share per year falls by 90 percent. Repairs have to be added to the account, along with new upholstery, refinishing, and any other work needed to keep the chair useful. If those costs are modest and continued use avoids replacements, several decades of service can be a powerful improvement. I did not need a millennium for that to matter.

Fifty years was a useful horizon to consider, not a number at which an environmental calculation told me to stop. Going from fifty to a hundred would halve the remaining annual share of the original manufacture. The question was how much extra material and work I should invest now for those more distant years, and whether anyone would want the chair long enough to get them.

Time complicates the accounting too. Environmental economists discount the monetary value of future damage, but that is not the same as making future emissions physically smaller. Keeping a chair in service postpones the need to make its replacement. By then, electricity or manufacturing might be cleaner. That possibility is different from applying a financial discount rate, and I was not counting on future improvements to justify waste today.

The years have to be useful ones. Hatnefer's chair survived millennia without providing millennia of seating. A chair I leave in a storeroom while buying another has not avoided that purchase. My father's rocker was a more relevant example: it was still doing its job, with occasional additions of wood, screws, and his time, then mine.

This brought the environmental question back to the design work. An accessible screw could make a small repair possible. A replaceable seat could keep damage to one part from ending the whole chair. Neither would help if nobody wanted to sit in it. The shape, comfort, and pleasure of ownership belonged in the same discussion as the materials and fasteners.

I wanted the old oak for reasons an environmental score would not capture. I liked saving scraps, learning a new process, and making something I wanted to live with. I did not have to turn each of those pleasures into a claim of environmental virtue. When I did make such a claim, I owed it a comparison. For these chairs, the opportunity I most wanted to pursue was a long working life.

My hundred-year and thousand-year ambitions could not go into the table as service already delivered. I could make the next repair easier and find out whether someone else could understand the chair without me. That would require handing over more than a photograph of a finished object.

A chapter of The Hundred-Year Chair, drafted and refined in public. The same text lives in the book; the interactive tools and the fully-sourced technical notes live here. home · the book · the technical notes · hundredyearchair.com