Additional Notes


One of the trickiest aspects of assessing a company’s level of responsibility is that no company is an island. In order to do business as usual, any company needs a steady influx of materials and services from other companies. A financial office needs paper to print on, power to run its lights, janitors to clean the floors, etc. Since these materials and services are an essential part of making a business’s end products, it is only fair that these materials and services factor into that business’s responsibility rankings. This is what we call branching.

With branching, when a company takes in products and services from other companies, it also absorbs at least some of these other companies’ responsibility data (RD – e.g., their worker pay, tons of pollutants created, etc). Let’s look at Jim’s Janitors, which cleans up at Claire’s Coffee Shop. Part of the pay for Jim’s janitors, the chemicals used to clean the floors, the amount of taxes Jim pays, etc, will all become a part of Claire’s rankings. Since Jim’s Janitors also cleans the floors and windows of forty other businesses too, though, it wouldn’t be fair to take all of Jim’s RD and factor it into Claire’s scores. Thus, the percentage of RD absorbed by any company A will always be equal to the percentage of company B’s total products and/or services that A is taking in. So for calculating Claire’s rankings, that means if 7% of Jim’s yearly billable man-hours are spent cleaning Claire’s coffee shop, then Claire’s business will absorb 7% of Jim’s RD.

Of course, to fully measure the environmental impact of those clean floors, we really must go deeper. What if, for example, Jim only uses mops made from old growth trees grown in rainforests. This biologically destructive purchasing habit means that by employing Jim, you’re encouraging (if only a small bit) the destruction of rainforests. Part of the simple beauty of branching, then, is that it keeps going back all the way to the source. Claire’s Coffee Shop absorbs part of Jim’s RD, but that’s only after Jim’s Janitors has absorbed part of the RD of, say, Mark’s Mops, who before that had already absorbed part of the RD of Rainforest Razers, the logging company that chops down those old growth trees. Thus, the CR Rankings of any one company include all of the hidden actors that contribute to that company’s final products, including the small bit of rainforest cut down thanks to Claire’s Coffee shop.

Such a seemingly tedious process might sound like worthless nitpicking at small potatoes, but every little bit matters. Those mops really are a small part of the effect of the espresso someone buys at Claire’s, so each caffeinated consumer should know what she’s supporting. Plus, since Jim’s CR rankings would drop thanks to those mops, he would be motivated to find different mops to buy. Losing such valuable mop customers would likely then also motivate Mark to find more sustainable wood sources, either helping to put Rainforest Razers out of business or convincing them, too, to find a more environmentally friendly way to do business.

What’s more, keep in mind that branching would quite often deal with much larger branches, such as the mops used in every Wal-Mart, or the manufacturers constructing every iPhone, or the shipping companies delivering all of Amazon’s products. Each seemingly small potato adds up to something quite big. Branching would give all companies, both big and small, an incentive to care about the responsibility of such hidden parts of their businesses, an incentive that should collectively make a huge impact on all facets of corporate responsibility.

Two-Way Branching

While branching would often be one-directional, it would also at times go two ways. When it came to Jim’s Janitors, for example, they provided a service to Claire’s Coffee Shop, and therefore made up part of the behind-the-scenes work needed to create Claire’s products. Claire’s business gave no service to Jim’s Janitors, though (unless Jim were to start stocking Claire’s coffee to pep up his janitors every day). Thus, Claire’s CR rankings absorbed some of Jim’s RD but not the other way around.

Sometimes, though, responsibility data would be shared in both directions, such as with a cereal company and a grocery store that sells that cereal. Just like the farmers harvesting the cereal’s grains, the store is an essential part of the cereal’s life cycle. Without the store, the cereal would never make it into anyone’s kitchen cabinet. On the flip side, though, the cereal (just like all other products the store sells) is a crucial part of the store. Without products to sell, a store would simply have nothing to offer. Therefore, each of the two companies would take some of the RD of the other in the same manner as stated before. If the cereal company’s products make up 2% of the grocery store chain’s yearly sales, then the cereal company would absorb 2% of the grocery store chain’s yearly RD. At the same time, if that grocery store chain sells 17% of the cereal company’s products for that year, then the grocery store chain would absorb 17% of the cereal company’s yearly RD.

The only catch with two-way branching is that anytime a company is absorbing RD, it would not absorb any redundant RD—that is it would not reabsorb its own data. Remember, for example, that the grocery store absorbs 17% of the cereal company’s RD. Remember also, though, that the cereal company as well absorbs 2% of the grocery store’s RD. So when the grocery store takes in the 17%, it could accidentally reabsorb part of its own data that was already factored into the RD of the cereal company. This kind of reabsorption, of course, would make no sense. It would turn into an endless feedback loop and distort the accuracy of CR Rankings. To avoid this scenario, whenever company A is absorbing RD from company B, any parts within company B’s RD that A has already absorbed will be considered redundant and blocked from reabsorption.

Corporate Responsibility Bureau

To effectively run the CR Rankings system, the government would need to establish a new, largely independent organization with centralized control over the rankings. While the exact naming and placement within the government are arguably not too important to us, we here recommend the Corporate Responsibility Bureau (CRB). The CRB would:

  • Collect all data needed for CR Rankings (with coordination from other government agencies like the IRS, OSHA, and EPA)
  • Calculate all rankings
  • Distribute rankings and labels to all applicable companies
  • Enforce the posting of rankings on all applicable products and storefronts
  • Publish a detailed account of all factors that went into the rankings of every company on a CRB maintained website
  • Investigate and prosecute cases of rankings fraud
  • Review and update data and metric setup as needed, so long as any changes are made public at least six months before taking effect

The CRB Board would be comprised of fifteen directors, each of whom would be voted into office by CRB Voting Members. Five directors would be voted in by voting members from each of three backgrounds, Workers, Environment, and Community. Those fifteen directors would then elect a president. Directors and presidents would serve six-year single terms, with no reelection.

CRB Voting Member status would be granted to any applicant who has worked in some capacity toward bettering corporate responsibility and/or bettering the world for our workers, environment, and communities for at least five years. Exact qualifications are subject to change (as to be decided by the CRB Board). However, such individuals shall include but not be limited to a(n):


  • Employee at any government agency that protects the rights and treatment of employees, such as the US Department of Labor
  • Member at any approved labor organization, such as the AFL-CIO, the National Education Association, the Service Employees International Union, etc.
  • Individual who otherwise actively works in some capacity to further the interests of employee protection and good treatment, be it for fair pay, safe workplaces, anti-discriminatory hiring, etc


  • Member of the National Academy of the Sciences
  • Employee at the Environmental Protection Agency
  • Individual who otherwise actively works in some capacity in the environmental sciences or to further environmental protection


  • Employees at and/or members of charitable non-profit organizations with no political or religious affiliations
  • Individual who otherwise actively works in some capacity to better the local community

Note that, while Congress could always pass laws that alter CRR, absent any such laws the CR Bureau would otherwise run and alter the CR Rankings program as it sees fit. CR Rankings would thus have the flexibility to deftly change as needed over time—to update its list of pollutants as new ones are created or discovered, to alter the data collection process to make it as comprehensive yet business-friendly as possible, to change the rules to eliminate any unfair loopholes discovered by certain businesses, etc.

Shadow Rankings

In addition to the rankings tabulated for each of CRR’s metrics (Distribution of Wealth, Carbon Footprint, etc), the system would also tabulate certain other shadow rankings that would not automatically count into a company’s official CR Rankings. This would be done for rankings that are useful in determining corporate irresponsibility (and thus for assessing any reason to lower rankings for Additional Factors metrics) but which, if made into full metrics themselves, would arguably create perverse incentives.

While shadow rankings could be much more extensive in the future, we currently include such rankings for:

  • Pay distribution by gender
  • Pay distribution by race/ethnicity
  • Pay raises by gender
  • Pay raises by race/ethnicity
  • Raises and firings for employees that report more work-related illnesses and injuries

To better understand shadow rankings, let’s look at gender pay. Because of the long history of pay discrimination in favor of men, gender pay and gender pay raise rankings would all be published and updated regularly online for all to see and would then give an unbiased, data-driven look at how companies fall in the spectrum in terms of pay equity. Using such rankings, the CR Bureau would then be able to much more reliably assess complaints of gender pay discrimination against specific companies and dock their Worker rankings accordingly with the Workers Additional Factors metric.

Of course, this begs the question: why not just make gender pay equity its own full, automatically calculated metric? Why do this as a shadow ranking and only take away points from certain companies? These are quite fair questions. A gender pay metric would no doubt be a perfectly just metric to have, given that wage discrimination is a very real and pressing issue of corporate irresponsibility, one that CR rankings should absolutely reflect and combat. At the same time, though, such a metric would unfortunately be very likely to create perverse incentives. For example, if such a metric ranked companies by average pay to each female employee, it would then reward companies that promoted and/or raised the pay of their female employees. However, it would also reward companies that do things like stop hiring new female workers. Why? Well, entry-level workers are generally paid much less than the average worker. Hiring new ones brings down the average pay of women overall. By the same financial logic, hiring more new entry-level men would also make the company seem more pay balanced. This could very likely drive down overall employment of women, an outcome that is obviously not the intent of such a metric and would therefore undermine its effectiveness and create a new, potentially worse problem.

If the metric instead just measured the total money paid to women versus the total amount paid to men, then a company could just hire 60% women but still give them lower-level jobs and pay them less each. Sound unrealistic? Well, look at the garment manufacturing and health care industries. Each is overall dominated by women, but the women are more skewed toward the lower-paying jobs like sewers and nurses, not higher-end jobs like managers, owners, and doctors. These industries would (inaccurately) look like models of pay equity with such a gender pay metric, though. Furthermore, such an automatic metric wouldn’t be able to take into account valid reasons that certain businesses might appear, based on the data alone, to be unfair to women. A family-owned restaurant might happen to be male-dominated simply because the members of the family happen to mostly be male. (A family-owned restaurant could also rank just as low because the family is made of mostly women.) In addition, a particular industry might not have many women because women have chosen not to apply there. The logging industry, for instance, has only 3.2% women,1 but is that because of discrimination or because few women find logging to be an attractive career path?

In other words, pay discrimination is, as far as we can tell, a hard issue to objectively rank with a data-based algorithm. The issue is better assessed with specific investigations into each company. That being said, though, shadow rankings would provide an invaluable tool to speed up such investigations and make them much more reliable and credible. When the CR Bureau receives plenty of complaints about the sexist lack of raises for women at Terry’s Textiles, the CRB can instantly access TT’s rankings. If the company does in fact rank low in these shadow rankings (i.e., it does pay its women poorly and gives them fewer raises than most other companies), then the CRB could then easily dock TT’s rankings.

Service Units

The service unit is a basic measure of what each company provides to the consumer. For food that means one kilogram of food sold. For energy that means one kilowatt-hour. For real estate that means housing for one person. Service units are essential to how CRR would calculate its Environment rankings. Without them, it would be quite tough to adequately assess the environmental impact of one company versus another.

Imagine, for example, comparing the carbon footprint of two oil companies. Our first instinct might be to compare the two companies by the total metric tons of carbon dioxide each company produces. Whichever company produces more tons of CO2 gets the lower ranking. Simple yet effective. Right? The problem here is the size of each company. One company may produce ten times as much CO2, but only because it’s ten times as big as the other company. It is no more environmentally unfriendly, but it would unfairly get a much lower Environment ranking. The second, almost-as-simple approach we would likely be tempted to use next to compare companies is by their environmental impact per overall revenue. This would at first seem to work relatively well. If that much bigger oil company makes ten times as much CO2, then we would divide its CO2 produced by its ten-times-as-high revenue. That would even it out compared to the much smaller company that divides by a much smaller revenue. We’re now much closer to a fair comparison, but the numbers would still be warped because this system would reward companies that charge more for the same service. Simply by doubling its price a company could greatly increase its Environment ranking, all without actually treating the environment any better. That doesn’t make any sense.

To best assess how well a company treats the environment, what we really want to compare is environmental impact per service provided to the consumer. This is really what we all already intuitively search for when we want to make such comparisons in our daily lives. If I want the more environmentally-friendly air conditioner, I want to know which one will use less electricity and fewer harmful chemicals per cooling of one house. I don’t care if the company uses more chemicals overall simply because it sells more AC units—I care how this one AC unit compares to another AC unit that I might buy. The same goes for those oil companies. What we really want is to compare the environmental impact of a gallon of gasoline from each company. In terms of calculating CR Rankings, this is where service units become quite necessary. Dividing a company’s environmental impact by the number of service units it sells gives us a much fairer, more accurate way of comparing how environmentally-friendly different companies are.

Part of creating CR Rankings would thus be creating and managing a comprehensive list of service units. When businesses report how much of each product it has sold each quarter, it would do so in terms of service units. A sample list of basic service units is listed below:


  • Raw materials
    : one kilogram
  • Food
    : one kilogram
  • Beverages
    : one liter
  • Electronics
    : one device
  • Computer software
    : one program or application
  • Textiles (clothing, bedding, curtains, etc)
    : one square foot
  • Transportation sales (e.g., cars or bicycles sold)
    : number of vehicles sold multiplied by average number of passengers found to use vehicle of that type
  • Real Estate (sales)
    : housing for one person
  • Energy
    : one kilowatt-hour


  • Shipping
    : one kg shipped one mile
  • Transportation service (e.g., commercial planes, trains, and busses)
    : one person taken one mile
  • Medical Care
    : one short-term ailment treated, or one long-term ailment treated for one year
  • Finance
    • One financial transaction facilitated (includes one money transfer, one ATM withdrawal or deposit, one check deposited or cashed, or any other transactions done at the behest of the customer; does not include any transactions done at the behest of the company, such as stock trades)
    • One dollar loaned for one year
  • Real Estate (rental)
    : housing for one person for one year
  • Cleaning
    : one square foot cleaned once
  • Food and Beverage Services
    : one person served
  • General services
    : one person served

This list of service units is a sample, as the entire list would be long and comprehensive, made to include all types of products and services legally sold within the country. The CR Bureau would be free to create, modify, or eliminate any service unit designations over time so long as the goal is always to more accurately and fairly compare the environmental impact of different companies.

One other very important note to make on the designation of different service units is that to achieve such accurate and fair comparisons, each service unit the CRB creates should focus on the consumer’s use of the good or service (i.e., the ends), not each business’s varying methods of delivering that good or service (i.e., the means). So if a person’s use of transportation is to get from point A to point B, we should therefore use the same transportation service unit of one person taken one mile regardless of whether it was a bus, plane, car, or bicycle used to get that person there. It might be tempting to create separate service units for cars and trains, for example, but that might unfairly reward the less environmentally friendly of the two (presumably cars). If cars were only ranked against other cars, not against more environmentally friendly transportation methods, a taxi company with the best gas mileage compared to other taxi companies could get a great Environment ranking. Meanwhile a bus company with a slightly lower gas mileage than other buses around would then get a low CR Environment ranking, even though taking that bus would surely have a lower carbon footprint than taking the best of the taxis. See the conflict there? The rankings would be backwards, all because we too narrowly defined each service unit type. So long as we define service units by the consumer’s use of goods and services, though, we should generally avoid that kind of conflict.

All in all, this is one of the trickiest aspects of creating CR Rankings, and one that would no doubt generate quite a bit of controversy. Allowing the CR Bureau the freedom to alter service unit classifications over time is essential to making the system as fair and effective as possible.

The Service Unit-Dollar

Simply by using standardized service units we can be assured that CR rankings will pretty fairly compare the environmental impact of two companies that provide the same kind of service (like two logging companies, two house builders, etc). But what if we want to compare two different types of companies (like a logging company versus a house builder)? This is where using the service unit alone breaks down. Any one service unit will inevitably have an arbitrarily set value that is different from all other service units. The average value of a kilogram of raw materials (like wood), for example, will be quite different from (and much lower than) the average value of housing for one person. If we were to so compare two such companies by their service units sold, the logging company would unfairly come out way ahead, simply because a kilogram of wood is obviously much smaller and easier to produce than an entire house, therefore presumably also requiring much less energy, pollution, and water to make.

Thus, what we will really use to compare environmental impact of any two companies is the service unit-dollar (SUD), i.e., the environmental impact per one dollar’s worth of service unit created. To calculate a company’s service unit-dollars, we multiply the number of service units produced by the average price of this service unit. Thus the general equation for calculating any of the environmental metrics will be:

Service Unit-Dollars

Environmental Impact

Or, a bit more specifically:

(Total service units sold) x (Universal mean price per service unit)

(Environmental Impact, per each metric)

By using the service unit-dollar instead of just the service unit, we account for the arbitrary difference in size and complexity of the average service unit of one type versus another. A company that builds houses would thus multiply the number of people it has housed by the average price of housing for one person. A logging company would multiply the number of kilograms of wood it produces by the average price of one kilogram of raw materials in general. We could then much more fairly compare the two companies, based on the environmental impact of one dollar’s worth of housing versus one dollar’s worth of lumber.

Universal Mean Price

One big question with this, of course, is why use the average price for each service unit (or “universal mean” as we have written above)? Well, first off, what this means is that if Dell sells, say, forty million computers one year, we would multiply that number by the average price of all computers sold in the US that year, not the actual price at which Dell sold them. So why don’t we just use Dell’s price? This distinction probably sounds awfully picky, but it’s important. To back up, remember that the reason we multiply by the price at all is so that we can compare the environmental impact of one dollar’s worth of service, not the impact of one service unit (since the size of each service unit is arbitrary). But if we were to use the specific price of Dell’s computers to accomplish this goal, Dell’s CR rankings could easily be skewed (and manipulated) based on how cheap or how pricey its computers were. So if Dell were to have the cheapest computers on the market that year, then it would have produced fewer service-unit dollars while still producing the same number of computers (and presumably the same amounts of pollution), thereby punishing themselves with a lower CR ranking. By artificially raising its prices, Dell could then game the system and increase its CR ranking for no legitimate reason (all while hurting customers with higher prices). Therefore, by using the average price of all service units of the same type sold that year (in this case computers), Dell has no artificial, bad incentive to raise its prices or to be punished instead for lowering them, all while we still get the benefits of fair environmental comparisons with the service unit-dollar.

Comparing Different Industries

Another question for the SUD, of course, is why try to compare such different companies at all? It’s inherently unfair, some might say. A computer manufacturer, a real estate business, and a logging company are so fundamentally different that it simply isn’t fair to compare them. If we’re going to compare companies at all we should only compare those in the same industry.

This is quite a fair objection, but there are two good reasons that we absolutely should compare all companies to each other. First, if we don’t compare all companies to each other, we would have to compare companies only within separate categories of industries. (All housing construction companies compared only to one another, all oil companies only compared to each other, etc.) The inherent flaw to this approach, though, is the overlap of different industries. If every company fit neatly into one category then this could potentially work. Tire manufacturing only, grocery sales only, website design only. But anyone in the business world knows that this isn’t the way things work. A large portion of companies, if not the majority, create products and services that overlap multiple industries. Netflix is a DVD rental service…turned internet streaming service, turned TV show creator. Honda makes cars…but also airplanes, boat motors, and robots. Trying to figure out whom else to compare these companies to would be an impossible nightmare. By using the service unit-dollar, though, all we have to do is add up each company’s overall SUDs (regardless of what types of sales they came from) and then divide by overall environmental impact. Done.

The second big reason to still compare different types of businesses is that we should know where each business and even each industry stands compared to all others. It’s probable that oil and gas corporations like ExxonMobil and Shell won’t do so hot in their Environmental rankings. Some in the industry would likely cry foul about this, that it isn’t fair. But if oil and gas production is an inherently dirty, environmentally bad business, shouldn’t we know that? If CRR consistently ranks oil and gas companies in the 1 to 2 range out of 10, that’s valuable information for consumers, and could potentially push those companies to support a move away from oil and gas. What’s more, if viable alternatives still don’t exist (like, say, electric power up stations for cars fueled by solar panels) then all of those gas stations with Environment rankings of 1 or 2 are really still just competing with each other. So if a 2 is the best that can be done in that field, then that will still attract more customers than a 1 (i.e., the only other Environment ranking seen in town at a gas station).

All-in-all using service unit-dollars to compare companies to one another gives the most accurate Environment rankings possible.

Innovation Points

One of the most difficult challenges we face today is that so many important new discoveries are never made because they aren’t deemed profitable. New drugs to treat diseases of poverty like Ebola. More biodegradable plastics. Safer methods for storing spent nuclear waste. Cheap water filtration systems for places without clean water. The list goes on and on.

To encourage such unprofitable innovations, the CR Bureau would award innovation points to any businesses that make these discoveries. Those points would then boost that business’s CR Rankings (by a specified amount for a specified period of time). Innovation points would thus create a profit motive for all kinds of breakthroughs that could do a huge, untold amount of good for the world.

There are two key stipulations for earning innovation points. The first is that such discoveries must make a big, positive impact on the world in a way that doesn’t already boost that company’s profits and/or CR Rankings. In other words, this mostly just pushes companies toward working on unprofitable discoveries—discoveries like cheap nutritional supplements to combat hunger, vaccines for tropical diseases like African trypanosomiasis, simple fixes to stop certain invasive species from destroying ecosystems, you name it.

There are, however, plenty of profitable discoveries—discoveries that help a company’s bottom line and/or boost its CR Rankings—that can still make the world a much better place. We should therefore still reward companies for such profitable discoveries with innovation points, so long as those discoveries are shared. The second key stipulation for receiving innovation points is thus exactly that. To get innovation points, each business would have to publish all details of its discovery to the public, so that its success could be easily and widely duplicated.

The issue here is that profitable discoveries are the kinds companies tend to be very reluctant to share. They would generally rather keep that information secret to maintain a competitive edge over their rivals. For instance, suppose a trucking company invents a new computer technology that reduces the fuel consumption of its truck fleet. The company could choose to keep that new technology private, cutting its own costs, boosting its Environment score with a lower carbon footprint, and then calling it a day. But what if that trucking company were to release this new fuel-saving computer program to the world, free for anyone else to use? That action could spark a huge industry-wide reduction in fuel use (and a corresponding reduction in greenhouse gas production). It could then spark even more innovation when, a few months later, a tech company improves on that computer program to make it even better. Because we want to encourage such sharing of information, CRR would award innovation points to the trucking company, not for the benefits it already enjoyed inside the company’s operations from its own lower use of gasoline (i.e., lower costs and a higher Environment ranking), but strictly for the benefits created outside of the company, that is the significant greenhouse gas reductions worldwide thanks to that new technology (i.e., benefits to the world that, without innovation points, would not have helped the company at all).

Bounty vs Reactive Innovation Points

Innovation points would be awarded in two ways. For bounty innovation points, the CR Bureau would set a bounty on a particular discovery that would fill an already known need. It’s the same kind of idea as the $10,000 bounty set for anyone who captures some notorious Wild West gunslinger. Here, the bounty would instead be a set number of innovation points to be given to the company that fulfills one of many needs we already know need to be fulfilled: cures for various diseases, green technologies we already seek to accomplish, etc.

Second, the CRB would award reactive innovation points to companies that develop unanticipated innovations that still end up benefiting the world. This could include all kinds of things, from an app that allows workers around the world to more easily report sweatshop conditions to a new power generator that uses a previously untapped renewable energy source. As with innovation points in general, though, reactive innovation points would exclude any innovations that enrich the world only for the sheer sake of having a better product (like Apple with the iPhone). Instead, they would be reserved for discoveries that help improve some sort of major problem for the world’s workers, environment, and communities, and only for the benefits to the world that aren’t directly helping the company in question. Obviously the line between the two is a bit subjective, which is why the CRB would make any hard decisions for what counts and what doesn’t.

Whether for bounty or reactive innovation points, the point level for each discovery would be set subjectively by the CRB according to how valuable it views that discovery to be for the world, levels that could be changed at any time to reflect the world’s changing needs.

Innovation Points Math

We already said that innovation points would raise a company’s CR Rankings, but let’s dive a bit deeper into how exactly. Innovation points would equate to a certain increase in rankings per million service unit-dollars (SUDs) sold by the company, or, really their average SUDs sold for each of the previous five years. Let’s unpack that. Let’s say the discovery of a new vaccine awards sixty innovation points. The CRB would have already set a points-to-rankings increase ratio for one million SUDs, which let’s imagine to be a 0.1 higher Community ranking for each twenty innovation points. That would then theoretically give a 0.3 bump for a company’s Community ranking for the company that discovers the vaccine. However, remember that this bump is really per million SUDs. That means the bigger the company is that achieves the innovation (i.e., the more SUDs it sells each year) the smaller the bump that company would get in its Community ranking for accomplishing the innovation. Why a smaller bump for bigger companies? Well, bigger companies pretty much by definition have bigger budgets. With bigger budgets, one can spend more on research and development. While we want big companies to aim for innovation points, we want small companies to do so, too. If the innovation points for the same discovery gave the same boost to a big company as to a small company, then the big company would have an unfair advantage in attaining innovation points and would gobble them up all the time. As with everything, the goal is to make a fair, competitive market, and by awarding each rankings increase per million SUDs, we make sure that all companies would have the incentive to strive for innovation points.

Preventing Fraud and Evasion

Any system encouraging more responsible behavior is only as strong as its enforcement mechanisms. As such, CRR would have strong punishments in store for companies that break the rules.

The main behavior CRR would presumably need to punish would be the falsification or omission of responsibility data. A business might intentionally fail to report the full number of hours driven by its company’s cars, for example, to try to get a higher Environment ranking. Much of the data CRR would rely on already has its own punishments in store for such false reporting (like jail time and fines for tax evasion), but CRR would add its own rather strict punishments for such crimes.

Enforcement Mechanisms

The first, perhaps most potent punishment for companies that are found to falsify their responsibility data is lowered CR Rankings. The CR Bureau would be authorized to subjectively lower the rankings of any company any amount based on any available evidence that the company in question tampered with the data it presented or failed to report certain data in order to get a higher CR Ranking. What’s more, any responsibility data (RD) first hidden but then uncovered by the CRB would then a.) be reincorporated into the company’s CR Rankings and b.) count for much more than it would have initially, multiplied by up to ten and counted into a company’s CRR for up to twenty years. Given that CR rankings are mostly based off of RD from the past year alone, getting caught trying to hide RD would make that toxic information count up to 200 times as much as normally, cleanly reported RD. That should be quite the incentive to just report the data honestly in the first place.

Falsifying and/or omitting RD would furthermore be considered a federal crime, punishable by fines, felony charges, and/or potential jail time. Repeated violations would also make a company subject to losing its business license within the country. Should that company be selling its goods from a foreign country, repeated violations would make the company subject to company-specific tariffs—tariffs that could rise in value as decided by the CRB given even more violations—or eventually even have the company’s products banned from sale within the country.

Catching Cheaters

In addition to having harsh punishments to deter cheating, CRR would also be structured to catch as many such cheaters as possible, should they try. First, much of CRR’s data reporting would be done with two-way verification. This data would have to be reported twice, in other words, once by each of the two sides in each transaction. For example, when Trisha’s Trucking reports the number of miles it shipped Bill’s Brooms this past year, it may be tempted to cheat and report more miles than it actually transported. That, after all, would increase the number of service units Trisha reports while keeping the environmental impact the same. Trisha could then see a higher Environment ranking. But on the flip side of this, Bill’s Brooms would have to report the number of miles the brooms were shipped, too. Bill wouldn’t want to have higher mileage listed, though. If anything he’d want to cheat by listing fewer miles than the brooms actually took to ship (for the same reason of getting a higher Environment ranking). Trisha and Bill are thus each wary of the other trying to cheat and highly unlikely to collude and cheat in the same direction, keeping the other in check. And if those two numbers are different when reported to the government, that discrepancy would automatically be red flagged for investigation. The same would go for the gasoline Trisha buys from Greg’s Gas. Trisha would want as low a number of gasoline gallons shown as possible, Greg would want as high a number as possible, and they would therefore not be motivated to cheat together. If either one cheats alone the two numbers will end up different, an easily caught discrepancy that would then be investigated.

Beyond two-way verification, CRR would have still more ways to catch cheaters. First, it would offer financial rewards to anyone who provides verifiable evidence of cheating, with bigger rewards for bigger instances of cheating. That would be reward money payable to, say, employees of the cheating companies, residents of the nearby town taking the brunt of the unreported pollution, or even rival companies sick of competing against unfairly false numbers. Note that because of the reward, all of these groups would have motivations opposite of the cheating company. Cheating companies would thus have a high likelihood of being reported, and any company thinking of cheating would thus have all the more reason to think twice before doing so.

One extra major asset the CR Bureau would have in catching cheaters is the vast trove of data it collects to run the CRR system. Part of its investigation of fraud would be to create algorithms that red-flag aberrant data, along with ranking the statistical unlikeliness of each bit of data provided. The more unlikely any RD reported, the higher up the priority list it becomes to be investigated.

With two-way verification, rewards for reporting data falsification, and sophisticated algorithms to catch fishy data, the CR Bureau would stand a high chance of catching any cheaters in the system. With the stiff punishments that follow, such cheaters would suffer greatly. Given the high risk of being caught and punished, most any business would have a strong motivation to never try to cheat the CRR system in the first place.

Work Done In Foreign Countries

Any company wishing to sell its products in a country that has enacted CR Rankings would have to report the same responsibility data. That goes for companies working anywhere around the world. So if the United States passes a law enacting the CRR system, any business anywhere in the world that wants to sell its goods in the United States would have to play ball and provide the same data to the US government—about how well it pays its workers, how much it pollutes, etc—regardless of whether its factory is in Kansas or New Delhi.

We discuss this universal nature of CR Rankings at much greater length in Our Current Approach is Doomed to Fail, but suffice it to say such a law would have much greater reach and power than localized laws that only govern behavior in the one small area where the law is passed. Also suffice it to say that a universal law like CRR has precedents before it and should operate entirely within a preexisting legal framework. No foreign business would have to sell its goods in a United States that runs the CR Rankings system, but if it chooses to do so then it would have to play by US rules—that is, to provide the data needed for CRR and print its CR Rankings on its products. It’s the same idea as how a foreign car company must abide by CAFE Standards in order to sell its cars in the US.

Of course, perhaps the biggest impediment to having foreign companies submit data even when their home countries have not enacted their own CRR system is that companies there could easily falsify data. Why send the real numbers of pollutants you’ve created when you could easily send fake, much rosier numbers instead? Your home government won’t punish you for it, so what do you have to lose? And who is around to catch your doing so?

The two main keys to prevent this kind of foreign data falsification are the CRR system’s direct punishments and market-based leverage. We list the punishments foreign companies would be subject to for fraud in the previous section on Preventing Fraud and Evasion: much lower CR Rankings, company-specific tariffs, and, in extreme cases, being outright banned from selling goods in the US. Catching foreign companies that do so may not be quite as easy as catching them here in the home country (if only because two-way verification may not be possible with some foreign companies), but rewards would still be given for credible information about rankings fraud, and algorithms would still be trained towards catching unrealistic data. Any other outside information gathered by the CRB would also factor into catching cheating companies, including investigations done by independent watchdog groups. The independent investigative nonprofit China Labor Watch, for example, reported in 2013 that Pegatron Shanghai, a factory that produces the iPhone, works its employees an average of six days a week for eleven hours a day at $1.50 an hour. After sending investigators into three Pegatron factories, CLW also reported 86 labor rights violations from underage labor to insufficient safety training to forced overtime.2 Such evidence would help further catch and punish foreign cheaters to make sure almost everyone plays by the rules and that those who don’t get punished.

The second key to having this universal system work, as we discuss in Our Current Approach is Doomed to Fail, is leverage. Enacting the CRR system in a country like the United States means enacting it where there is also the largest consumer market in the world. Nowhere else in the world has as lucrative a pool of customers as the US. The threat of low rankings, tariffs, or being completely banned from the fattest, juiciest marketplace in the world is therefore quite the potent threat. This gives CRR leverage over foreign companies—leverage to keep them from cheating. The same would go should CRR be enacted in any other high-consumption countries, like Australia, Canada, Japan, the UK, and European Union member states. In any such country, foreign companies would have a lot of lucrative sales to lose should they choose to cheat.

Of course, talking of enacting CRR in other countries brings us to the ideal end-goal scenario, which is to have CRR passed as the law of the land throughout the world. The more countries that sign-on to the system, the more easily all countries can help police the businesses within their own countries to ensure honest cooperation.

Part-Time Workers

In addition to the seven aforementioned questions to be asked as part of the yearly individual federal tax return and used to calculate Worker Safety & Health scores, each worker would also be asked the following eighth question: “If you work part-time (i.e., fewer than 35 hours a week) do you do this by choice or because your employer will not offer you full-time employment?” For calculation of Workers subsections “Distribution of Wealth” and “Pay Relative to Local Standard of Living,” employees who answer that they work part-time by choice will have their total yearly compensation divided by their total yearly number of hours and multiplied by the median yearly number of hours of all employees at the company in question who either work full-time or work part-time but not by choice. By calculating these metrics in this way, companies who hire such by-choice part-time workers will not be punished with abnormally low yearly compensations for such workers. Those part-time workers who mark that they work part-time not by choice, on the other hand, will have their yearly compensation kept as is for calculation of the above metrics. In such a situation the company will be responsible for the low yearly pay of part-time employees, thus encouraging the company to create jobs that help its employees live comfortably year-round.


Internships at for-profit companies are to be considered, by default, employment positions, such that whatever wages (or lack thereof) given to such interns will be counted just like the wages given to all other employees in the calculation of all CR Rankings. (In other words an unpaid internship should be an actual internship, not an excuse for free labor.) In order to have an internship be discounted from consideration for a company’s CRR, the company in question must provide clear evidence that such internships pass all six of the following “unpaid internship” requirements (as issued by the US Department of Labor):

1. The internship, even though it includes actual operation of the facilities of the employer, is similar to training which would be given in an educational environment

2. The internship experience is for the benefit of the intern

3. The intern does not displace regular employees, but works under close supervision of existing staff

4. The employer that provides the training derives no immediate advantage from the activities of the intern, and on occasion its operations may actually be impeded

5. The intern is not necessarily entitled to a job at the conclusion of the internship

6. The employer and the intern understand that the intern is not entitled to wages for the time spent in the internship

This requirement still allows companies to provide unpaid internships when the internships actually do function as internships. When instead these internships are just an excuse for free labor, the company in question will be punished with a lower Workers CR Ranking.

Makeup of CRB Committees Ruling on Discrimination

Whenever a CRB Committee would be commissioned to rule on changes in a company’s rankings stemming from discrimination within the Additional Factors metrics, the committee in question would (if at all possible) have at least a quarter of its members come from the group of people potentially discriminated against, if not more. So, for a pay discrimination case regarding African-Americans, the committee should be at least 25% African-American, etc. This would be done in order to assure fairness to the potentially aggrieved group.

Not-For-Profit Companies

Non-profit companies will still be ranked in CRR but would have several differences from normal, for-profit companies. They are as follows:

  • Shouldering the Tax Burden
    : Not applicable to non-profits, since they do not pay taxes
  • Charitable Giving
    : Not applicable to non-profits, since they are the very beneficiaries of charitable giving. It would fundamentally not make sense to encourage them to re-donate their donations to other charitable organizations.
  • Environment (all)
    : Each metric would be calculated as the environmental impact per one dollar of budget used (not per SUD). Many if not most non-profits have goals that are too vague to easily assign them service units (like generating awareness, lobbying politicians, etc). Even such services that could be defined and categorized with service units, though, have no discernable market value since such services are not for sale. Hence trying to use the normal system of service unit-dollars does not seem feasible for non-profits, so instead we use the second best option of environmental impact per one dollar of budget spent.

Basic Needs

For the calculation of the Local Standard of Living (for the Pay Relative to Local Standard of Living metric), if median housing, utilities, food, and health care costs for a locality do not already ensure adequate access to safe housing, clean water, sanitary waste disposal, the fulfillment of basic caloric needs, a K-12 education for all children in the family, and reliable internet access, then the gap between the average estimated cost of fulfilling any of these basic needs and the actual median cost of living will be added to the Local Standard of Living monetary sum for each county/city. This should ensure that companies don’t automatically get rewarded for paying its employees well relative to the local median cost of living…if that local median cost of living is so low that it still spells poverty for the locals living at median levels of wealth.

Eliminating Statistical Outliers

For each set of relative 0 to 10 rankings given for each metric, the top two percent and the bottom two percent of companies will all be given a 10 and a 0, respectively. When determining the relative rankings of all other companies in between, the 98th percentile will be used as the benchmark 10 score and the 2nd percentile will be used as the benchmark 0 score. This will be done in order to eliminate the influence of any statistical outliers. Without this change, one issue is that certain companies could feasibly try to undermine the rankings system by creating a separate shell company with absurdly high or low raw CRR results (e.g., it only pays two people and pays them equally for a perfect Workers ranking). Getting artificially low results for this fake company could make its own low ranking much higher by comparison, or getting artificially high results could then bring its elite competitors to lower rankings by comparison. Such tampering wouldn’t be good for anyone. Setting the bars at the 2nd and 98th percentiles should do much to eliminate the threat of such meddling.

Individualized Metric Percentages

One of the wonderful strengths of our world is its diversity of opinions. It should therefore come as little surprise if, upon the enactment of CR Rankings, some individuals out there might disagree with the exact percentages given to each metric within the system. Some might think that Distribution of Wealth should matter the most. Others might say Shouldering the Tax Burden. Others still might side with Non-Greenhouse Pollution as the most important factor in judging each company.

One thing to note is that all metric percentages would be subject to change as the CR Bureau sees fit, allowing for popular changes to how the metrics are weighted in a way that would likely satisfy many if not most citizens who disagree with the current setup. Should any individual consumer still not be satisfied with these percentages, though, all of the data the CRB collects and the rankings created from them would be 100% available to the public. Therefore, anyone could foreseeably create his or her own metrics percentages. It should be quite feasible to create phone apps and websites that would allow each consumer to create their own metrics percentages and then rank all companies accordingly.

A System Built To Change

What we put forth on this website is the beginning vision of how to structure the CRR system. We have toiled over every detail of its construction to make sure this system is as simple, efficient, and effective as possible. We thus feel quite confident that the CRR system, as described, would work incredibly well and would bring widespread positive change.

That being said, however, Corporate Responsibility Rankings is a system built to change. The world and our understanding of it constantly changes, and as a result, CRR can and almost definitely should change over time too. Once the law is enacted, the CR Bureau would be in charge of making sure CRR adapts as needed to always best fulfill its purpose.

The CRB Board would meet at least once every six months (if not more) to alter any rules as they see fit. When evidence of previously unknown pollutants surfaces, then such pollutants should be added to the list of pollutants that factor into a business’s Environment score. If climate change comes to look even worse over time, then perhaps it should be adjusted to take up more than 40% of the Environment ranking. On the other hand, if someday down the road we get our greenhouse gases under control, then perhaps Carbon Footprint could come down below 40%. Metrics could be added. Metrics could be deleted. Overall the CRB should do whatever is needed to update these rankings over time so that they may always reflect our best idea of how to most accurately measure corporate responsibility and how to best motivate companies to do better.

Even now before this system is put in place, though, we welcome any suggestions for how to make the system work better. Over the coming months and perhaps years we will most definitely fine-tune the system based on your ideas and on any other evidence showing that it could be even better structured. If you have any such feedback to give now, please do at our Contact Us page!


1 “Employed Persons by Detailed Industry, Sex, Race, and Hispanic or Latino Ethnicity.” Labor Force Statistics from the Current Population Survey, US Bureau of Labor Statistics, 8 Feb. 2017, Accessed 7 Sept. 2017.

2 Apple’s Unkept Promises: Investigation of Three Pegatron Group Factories Supplying to Apple. China Labor Watch, 29 July 2013,