And this may be essentially the same tack China has taken with the IP issue. While various efforts to tighten IP rights enforcement have been publicized, it's not clear how heartfelt this effort has been. Research by the Business Software Alliance and IDC, for example, reveals that China's use of illegal software in 2009 amounted to a commercial value of $7.6 billion, a $900 million increase from the previous year (the largest increase of any nation).

The truth is that China is not working hard to protect IP, but the conventional wisdom in the west is that, once China's own economy begins to create intellectual property with global value — patents, drug formulations, movies for the world market — its government will see more gain than pain in a stringent IP regime.

But is that right to assume?

Think about it: If the Industrial Economy died and made you God (or China), what kind of IP rights would you design for the new, clean slate information economy? One knee-jerk response is to favor the same kind of regime. "Since information is so valuable," you might say, "it must be protected with even more Draconian rules." But the industrial paradigm isn't a match for the information economy. The crucial difference, utterly fundamental but only in the early stages of being worked out, is the zero cost of reproduction of information goods.

You must know this already, but let's look straight at it: if a farmer produces a bale of hay, one horse or another eats it, but not both. A steel mill's ingot goes into a sedan or a skyscraper, but not both. So a price mechanism and market is needed to mediate the competition for a scarce resource. But when a hacker produces a new capability on Linux, any number of people can use it without taking it away from anyone else. We can all have our code and eat it, too.

Given that reality, should SAP try to get Germany to ban the distribution of Linux code?
For an economist of the orthodox sort, this does not bear worrying about, because Linux must be a figment of the imagination. If the hacker isn't being paid, there is no economic incentive to produce it, and it will never happen. Yet it does happen, because the hackers take their pay in other currencies. (See Yochai Benkler's work for more on the rewards of "collaborative production.")

So let's look at a corporate case. The world's drug companies have a trove of accumulated research. The successful research is disclosed as part of the licensing process. The research leading to dead ends, however, sits alongside the Ark of the Covenant in the warehouse from Indiana Jones and The Raiders of the Lost Ark. In your Godlike view, which would be better for the economy, keeping it there or disclosing it to the world of researchers? Certainly there would be cases in which one scientist's ceiling would become another's floor, and a recombination of knowledge would accelerate progress — once again at no marginal cost. And if you add non-economic considerations like, say, the benefits of accelerated drug discovery, there's a huge welfare gain to the world's population. In a society truly based on information economics, the non-disclosure of drug research might properly be prosecuted as stealing.

The oft-cited reason to crack down on free taking of information is that limiting it is necessary to incent production. Yet the free sharing of music files has hardly stopped people from making music. In fact, music has never been a rational economic choice of profession. Paying iTunes, or bands, for music you could download for nothing isn't "rational" either. But in that corner of the economy, all the players now seem to be working out new and appropriate models. And at least we've learned enough that no one is trying to sue iPhone app developers who give their work away for free.

What about patents? Lawrence Lessig has been the definitive voice arguing that the supposed need for patents to provide incentives is a rationale, not a fact. In a world in which the competition is someone not copying you but obsolescing you, protection becomes less and less important. But in any case, we're not worried about maximizing anyone's profits or competitive advantage — we're trying to devise the economic system that works the best for the society. In Regional Advantage, Annalee Saxenian analyzed the competition between Silicon Valley and Massachusetts' Route 128 for leadership of the high-tech economy during the 1980s and 1990s, and concluded that openness, both through explicit contact and circulation of people, accounted for Silicon Valley's triumph.

This is not an argument for doing away with all patents and licenses, only that their availability should be mandatory on some reasonable terms. Creative Commons is an attempt to provide a set of terms suitable for non-fiction authors, for example.

We are still in the horseless carriage stage of the information economy. That is, we've learned how to use new devices that substitute for and add capability to what we had before, but we haven't reoriented our thinking to organizing our economy around the characteristics of the new technology. Just as the functional organization, the DuPont Equation, standard costing, and the methods of 20th Century management arose first in the innovative industrial organizations like DuPont and GM, the appropriate rules are appearing first in information-based organizations like Linux or Facebook. Reputation, acknowledgement, privacy, and accessibility are among the dimensions that matter here, as direct cost, scale, and productivity did in the past.

Executives have snorted derisively at the "gift economy" as economists sneer at Gross National Happiness. But these attitudes are founded on a deep belief in scarcity as the problem economies have to solve. The economy in which this belief is warranted isn't disappearing — we still need satellites, fibers, and silicon foundries to move that free information. Almost by definition, economists have trouble counting things that don't have costs and prices. And in the industrial world, too low a price leads to overconsumption. Excesses like the overuse of fossil fuels result. But what would it mean to over-consume knowledge?

How the two sectors — the information economy and its industrial infrastructure — will interact is subject for another day. But the next decade, as foretold by the Benklers , Saxenians, and Lessigs, will see a shift from protecting the old IP regime to figuring out the new.

Of course, China might still change its mind — the first global blockbuster film made in China was released to the world market last month. But I bet there were as many camcorders in the theater on opening day as there were cameras at Expo last month.

And this may be essentially the same tack China has taken with the IP issue. While various efforts to tighten IP rights enforcement have been publicized, it's not clear how heartfelt this effort has been. Research by the Business Software Alliance and IDC, for example, reveals that China's use of illegal software in 2009 amounted to a commercial value of $7.6 billion, a $900 million increase from the previous year (the largest increase of any nation).

The truth is that China is not working hard to protect IP, but the conventional wisdom in the west is that, once China's own economy begins to create intellectual property with global value — patents, drug formulations, movies for the world market — its government will see more gain than pain in a stringent IP regime.

But is that right to assume?

Think about it: If the Industrial Economy died and made you God (or China), what kind of IP rights would you design for the new, clean slate information economy? One knee-jerk response is to favor the same kind of regime. "Since information is so valuable," you might say, "it must be protected with even more Draconian rules." But the industrial paradigm isn't a match for the information economy. The crucial difference, utterly fundamental but only in the early stages of being worked out, is the zero cost of reproduction of information goods.

You must know this already, but let's look straight at it: if a farmer produces a bale of hay, one horse or another eats it, but not both. A steel mill's ingot goes into a sedan or a skyscraper, but not both. So a price mechanism and market is needed to mediate the competition for a scarce resource. But when a hacker produces a new capability on Linux, any number of people can use it without taking it away from anyone else. We can all have our code and eat it, too.

Given that reality, should SAP try to get Germany to ban the distribution of Linux code?
For an economist of the orthodox sort, this does not bear worrying about, because Linux must be a figment of the imagination. If the hacker isn't being paid, there is no economic incentive to produce it, and it will never happen. Yet it does happen, because the hackers take their pay in other currencies. (See Yochai Benkler's work for more on the rewards of "collaborative production.")

So let's look at a corporate case. The world's drug companies have a trove of accumulated research. The successful research is disclosed as part of the licensing process. The research leading to dead ends, however, sits alongside the Ark of the Covenant in the warehouse from Indiana Jones and The Raiders of the Lost Ark. In your Godlike view, which would be better for the economy, keeping it there or disclosing it to the world of researchers? Certainly there would be cases in which one scientist's ceiling would become another's floor, and a recombination of knowledge would accelerate progress — once again at no marginal cost. And if you add non-economic considerations like, say, the benefits of accelerated drug discovery, there's a huge welfare gain to the world's population. In a society truly based on information economics, the non-disclosure of drug research might properly be prosecuted as stealing.

The oft-cited reason to crack down on free taking of information is that limiting it is necessary to incent production. Yet the free sharing of music files has hardly stopped people from making music. In fact, music has never been a rational economic choice of profession. Paying iTunes, or bands, for music you could download for nothing isn't "rational" either. But in that corner of the economy, all the players now seem to be working out new and appropriate models. And at least we've learned enough that no one is trying to sue iPhone app developers who give their work away for free.

What about patents? Lawrence Lessig has been the definitive voice arguing that the supposed need for patents to provide incentives is a rationale, not a fact. In a world in which the competition is someone not copying you but obsolescing you, protection becomes less and less important. But in any case, we're not worried about maximizing anyone's profits or competitive advantage — we're trying to devise the economic system that works the best for the society. In Regional Advantage, Annalee Saxenian analyzed the competition between Silicon Valley and Massachusetts' Route 128 for leadership of the high-tech economy during the 1980s and 1990s, and concluded that openness, both through explicit contact and circulation of people, accounted for Silicon Valley's triumph.

This is not an argument for doing away with all patents and licenses, only that their availability should be mandatory on some reasonable terms. Creative Commons is an attempt to provide a set of terms suitable for non-fiction authors, for example.

We are still in the horseless carriage stage of the information economy. That is, we've learned how to use new devices that substitute for and add capability to what we had before, but we haven't reoriented our thinking to organizing our economy around the characteristics of the new technology. Just as the functional organization, the DuPont Equation, standard costing, and the methods of 20th Century management arose first in the innovative industrial organizations like DuPont and GM, the appropriate rules are appearing first in information-based organizations like Linux or Facebook. Reputation, acknowledgement, privacy, and accessibility are among the dimensions that matter here, as direct cost, scale, and productivity did in the past.

Executives have snorted derisively at the "gift economy" as economists sneer at Gross National Happiness. But these attitudes are founded on a deep belief in scarcity as the problem economies have to solve. The economy in which this belief is warranted isn't disappearing — we still need satellites, fibers, and silicon foundries to move that free information. Almost by definition, economists have trouble counting things that don't have costs and prices. And in the industrial world, too low a price leads to overconsumption. Excesses like the overuse of fossil fuels result. But what would it mean to over-consume knowledge?

How the two sectors — the information economy and its industrial infrastructure — will interact is subject for another day. But the next decade, as foretold by the Benklers , Saxenians, and Lessigs, will see a shift from protecting the old IP regime to figuring out the new.

Of course, China might still change its mind — the first global blockbuster film made in China was released to the world market last month. But I bet there were as many camcorders in the theater on opening day as there were cameras at Expo last month.

In our increasingly multi-polar world characterized by the inexorable rise of emerging markets like Brazil, China, and India, I believe multinationals (MNCs) must abandon their ethnocentric innovation model — which concentrated all their R&D resources in the West. Instead, MNCs must embrace a "polycentric" innovation model in which R&D capabilities are distributed globally to swiftly seize regional opportunities and yet are integrated into a loosely-coupled global innovation network to drive creative synergies on an international scale.

This polycentric, networked innovation model is vital for MNCs if they wish to succeed in emerging markets like India. Why? Because as Shiv Shivakumar, CEO of Nokia-India points out: "Merely thinking of growth happening in emerging markets won't help a MNC grow in those regions." Translation: these markets won't emerge out of nowhere. The onus is on MNCs to first shape emerging markets before they could profitably serve them.

The best way for an MNC to shape (and lead) a market is by building up more local R&D capabilities and cultivating a vibrant local partner ecosystem so it can systematically design and market locally-relevant offerings. But that requires shifting MNCs' center of R&D gravity from the West to the East.

But polycentric innovation isn't just about scaling up your R&D operations in countries like India and China only for the sake of serving local markets. The next step after that is to import frugal innovations like mobile banking services first deployed in emerging markets into home markets in Western nations where consumers — still reeling under the recession — are clamoring for affordable yet high-value goodies. As such, I see an MNC that adopts the polycentric innovation model evolving through four successive stages of maturity:

• Stage 0: At this stage, the MNC's R&D operations are mostly concentrated in the West. While the MNC operates in emerging markets, their local units are primarily sales and marketing functions.
• Stage 1: Here the MNC starts shifting some of its R&D work to low-cost countries like India that offer plenty of high-quality scientists and engineers. However, this local talent is primarily used to design and develop solutions for Western markets. The Indian IT outsourcing phenomenon epitomizes Stage 1.
• Stage 2: The MNC recognizes the massive potential of emerging markets and delegates more responsibilities to local units in emerging markets which initiate and manage their own R&D projects to cater to local needs. But the P&L responsibilities for new product lines launched in emerging markets remain held by senior execs located in Western headquarters.
• Stage 3: The MNC starts networking R&D activities in emerging markets with the rest of their global network in order to cross-pollinate ideas for new products and business models across multiple regions. The P&L responsibilities for new product lines gradually shift to emerging markets.
• Stage 4: At this ultimate stage, the R&D hubs in emerging markets are given a global remit as they now own the P&L responsibilities for the global design and rollout of new products. Senior execs physically located in India and China are now in charge of global business units and entire divisions.

Having closely studied the innovation models of several multinationals in recent months, I would say that 30% of Fortune 500 firms are still 'stuck' in Stage 0, while 40% have already leaped into Stage 1, while 20% have braved their way into Stage 2. As such, I believe that only 9% of multinationals are at Stage 3, whereas merely 1% or less has made it all the way to Stage 4. I would peg Microsoft to Stage 2 and IBM to Stage 3, while Cisco and Nokia are comfortably sitting in Stage 4.

Where's your firm on this continuum?

Speaking of the shift of (R&D) gravity from West to East, I will be attending the World Economic Forum's India Economic Summit in New Delhi on November 8-10. This Summit, which marks the 25th year of the World Economic Forum's engagement in India, is themed "India's Next Generation of Growth." During those three days I will be blogging daily on his site about the exciting business and socio-cultural trends that are reshaping India in the 21st century. Hope to see you here!

There is an overlap between reverse innovation and disruptive innovation but not a one-to-one relationship. In other words: Some, but not all, illustrations of reverse innovation are also illustrations of disruptive innovation.

A reverse innovation, very simply, is any innovation likely to be adopted first in the developing world. It is so called because historically nearly all innovations have been adopted first in rich countries. We argued that reverse innovation will become more and more common, and that it presents a formidable organizational challenge for incumbent multinationals headquartered in the rich world. We also explained an organizational model for overcoming that challenge.

A disruptive innovation has a particular dynamic that endangers incumbents. The incumbent's product has two primary dimensions of merit, A and B. (For example, A could be quality and B could be speed of delivery.) Mainstream customers are mostly interested in A but there is a minority customer set that values B more than A. The disruptive innovation, at launch, is weak on A but strong on B. As such, it attracts only the minority. Because mainstream customers don't want it, incumbents tend to ignore the new entrant and the new technology. But over time, technology improves, and the innovation gets better and better at A. Eventually it meets the needs of mainstream customers on dimension A, and, since they also place at least some value on B, they start choosing the new product. The incumbent is suddenly disrupted; they have ignored the new technology all along.

In Christensen's famous study of the disk drive industry, A was the capacity of the disk drive and B was the size of the disk drive. Christensen showed that new entrants repeatedly disrupted incumbents by introducing smaller disk drives with lower capacity. Initially, mainstream customers were uninterested. They needed more memory, not less. But, over time, the capacity of the smaller drives went up and up until mainstream customers were interested.

So, what is the relationship between the reverse innovation and disruptive innovation? We see three primary situations that create the possibility of reverse innovation. Only the first is also an illustration of disruptive innovation.

The first is created by the income gap between rich countries and developing ones. Because per-capita incomes are so low in the developing world, conditions are ripe for innovations that offer decent quality at an ultralow price — that is, a 50% solution at a 5% price. At first, the 50% solution is unattractive in the rich world, but eventually, performance rises to the point that it is attractive in the rich world. This is clearly also a disruptive innovation story, where A is performance or quality and B is price.

The second is created by the infrastructure gap between rich countries and developing ones. Most of the infrastructure (energy, transportation, telecom, and so forth) in the developing world has yet to be built. As such, demand for new infrastructure technologies is much higher in the developing world than it is in the rich world, where demand for infrastructure is created primarily by the need to replace existing infrastructure. This is not an illustration of disruptive innovation.

The third is created by the sustainability gap between rich countries and developing ones. Many developing nations are confronted with environmental constraints far sooner in their path of economic development than rich nations were. Desalination technologies, for example, are likely to be adopted in places like Northern Africa before the desert southwest in the United States needs them. This is also not an illustration of disruptive innovation.

Whether an innovation is reverse, disruptive, or both, it is difficult for an established organization to execute. For reverse innovations, companies must overcome resistance to shifting power and control away from headquarters, and they must be willing to reshape the organizational models and expectations of in-country teams. For disruptive innovations, companies must overcome the initial resistance to prioritizing an investment that does not interest mainstream customers. And, even if they do invest, they must overcome the fear that the new product will eventually cannibalize the existing business.

Vijay Govindarajan is the Earl C. Daum 1924 Professor of International Business and director of the Center for Global Leadership at the Tuck School of Business at Dartmouth and is professor in residence and chief innovation consultant at GE. Chris Trimble is on the faculty of Tuck and consults to GE.