« Love in the Time of No Time |
Main
| Communication vs Gadgets »
November 24, 2003
Social Capital as Credit
Posted by Ross Mayfield
Social capital, or aggregate (connected) reputation, is a form of credit. Some formal transactions can be supported by social capital. Informal transactions are rarely underpinned by financial credit or legal agreement and instead rely entirely social capital. We all have our internal calculators keeping tacit track of who is wronging and righting, the health of the relationships and adjusting our actuarial tables according to experience.
Sometimes a service arises to make a portion of this explicit, such as Ebay's or Slashdot's reputation system. Scaled reputation systems to date are, in fact, subjective proxies – a set of localized decisions that result in a visible emergent pattern – the pattern itself being open to interpretation. But if there is enough social agreement to play the rating game, they decrease transaction risk and increase liquidity. Not just for those with better ratings, but for the network or market as a whole.
Say you are selling a couch to a neighbor at a garage sale who is going to pay for it when its delivered next weekend. The transaction could be supported by formal credit using means as varied as an IOU or contract (providing legal recourse as credit) or financial means such as a deposit, escrow or credit card. It could also be supported by social capital. The key difference is an implied agreement that default means a negative impact on the defaulting party, not from explicit penalties, but to reputation with the seller and others in the neighborhood that is a network. The overhead cost for securing a transaction with financial credit is greater.
Using social capital to underpin transactions is an iterative approach. It only works if there will be future transactions and each occurs within the context of a social network. Game theory holds the best strategy is to trust but verify with each iteration. This presents greater risk up front, but builds trust and reputation with each iteration, so over time transaction risks and costs decrease. But it should be clear that without additional legal recourse for default it only works for smaller transactions.
In absence of formal credit, social capital is the norm. Micro-markets, more traditional cultures and third world countries practically run on social capital as a result. Up until the advent of the Internet, markets and networks that run on social capital were unable to scale. The sad irony is that the markets that need scaled reputation the most still lack access to supportive technologies.
The only cap to abundant potential connections is our mental capacity to manage relationships (150 active ones at a time). New tools are giving us greater capabilities to recall and invoke latent ties, but this is a hard barrier. What’s interesting is how with the cost of group forming falling, local networks are becoming denser, membership more dynamic and new clusters of localized decisions are ripe for enabling emergent patterns.
The potential supply of social capital is abundant, only held back by search and transaction costs. Social software and social networking are rapidly driving these costs towards zero. The pace of capital formation is accelerating because of two additional factors.
In the parlance of network or systems thinking: in the absence of connections, nodes become state attractors. In other words, when the amount of connections is limited, the value of connections is high.
Economists have an applicable rule for this as well:
Say’s Law, or “supply creates its own demand.” Now Say’s Law doesn’t work when there is money involved (creates an arbitrage opportunity, otherwise supply-side economics would make sense), but it does apply to barter, reputation and micro-markets. When money is involved, it provides a universal arbitrage path, causing a fight over equilibrium and discounting the impact of Say’s Law at a macro scale.
This is one reason why you can’t trade goods or cash for social capital. Or if you do, it disrupts equilibrium across markets. Now I am sure some elaborate schemes have allowed traders on eBay to assume others’ identities and some virtual world economies have crossed this boundary. But the point is you can’t monetize social capital in aggregate, because it operates at a micro-scale. You can foster social capital for the value of its emergent patterns and what it enables: the flow and production of other tangible and intangible assets. The value of social capital is local, but its impact is global.
Comments (11)
+ TrackBacks (0) | Category:
- RELATED ENTRIES
- › Spolsky on Blog Comments: Scale matters
- › "The internet's output is data, but its product is freedom"
- › Andrew Keen: Rescuing 'Luddite' from the Luddites
- › knowledge access as a public good
- › viewing American class divisions through Facebook and MySpace
- › Gorman, redux: The Siren Song of the Internet
- › Mis-understanding Fred Wilson's 'Age and Entrepreneurship' argument
- › The Future Belongs to Those Who Take The Present For Granted: A return to Fred Wilson's "age question"
1. Will Davies on November 25, 2003 3:29 AM writes...
What you're describing may be one form of social capital, but it is a fairly perverse form. To see social capital as a form of currency makes sense by analogy to digital reputation systems, but is rare elsewhere. I wonder if you're assuming that social capital is nothing more than trust in exchange situations?
Far more commonly, social capital has little directly to do with exchange. When you join a sports club, you do so because you want to join a sports club. You may happen to discover some new job opportunity in the process, or find someone to buy your car - but that's not why you joined the sports club.
When middle class parents and grandparents chat with their children, they pass on a form of cultural and social advantage that is social capital (the original definition of social capital). This will support the economic chances of that child through life, but its hardly a form of credit that has been handed over.
Describing communities and networks as 'capital' doesn't necessarily mean that we think of them as 'capital'. But it may lead us to over time. Throw in codified reputation systems, and you start to get a society such as you describe, where favours are accounted for and reputation calculated. This may be an effective society, but it sounds a fairly grim one.
Permalink to Comment2. Ross Mayfield on November 25, 2003 11:53 AM writes...
Absoultely and well said. I am describing one of many facets of social capital, as aggregated reputation or trust, which I don't believe has been covered.
My other point is the internal calculations are not made explicit and especially do not scale beyond local groupings. When done as codified reputation systems, they are merely proxies or games we play and not actual valuations. The notion of Whuffie needs to be kept in check with what it would represent.
Permalink to Comment3. phil jones on November 25, 2003 11:56 AM writes...
I was struck by your "third world" comment.
Hernando De Soto (http://www.amazon.com/exec/obidos/ASIN/0465016146/thoughtstorms-20/002-7811354-2640007) basically argues that the third world already has perfectly good social mechanisms to safe-guard property, but lacks the higher-level, more abstract, legal conception. And this is what keeps it poor.
I don't know if I agree with him, but I wonder whether there's a case that social capital lacks the kind of flexibility of more artifical legal constructs (as well as your point that it doesn't scale.) Will that always keep it in a minor role?
Permalink to Comment4. Ross Mayfield on November 25, 2003 2:02 PM writes...
Social capital plays a major role in the absence of legal constructs. And while its use is inherently local its impact is global.
People are innovating on the basis of social capital augmented by networked technologies. Open source is the prime example. People make decisions to produce largely within localized (and I don't me geography) social context. The pattern that emerges, a new model of production (commons-based peer production), results in contructs of global utility. Open source is the result of the earliest adoption of social infrastructure, what's happening now with weblogs and media/politics seems to be next.
In Francis Fukuyama's Trust he talks of two kinds of nation-states, institution-centric and family-centric. For the first and second world, institutional constructs are arbiters of trust. In the third world, trust is more closely held. Economic theory explains up to 80% of economic events, which has centered on individual and organizational incentives. The remainder is the relm of relationships. Seems we are still at the stage of developing frameworks and tools to understand these dynamics.
What's interesting to me is how globalization threatens the family-centric model and technology can bridge relationships across boundaries of family and state. This runs in direct counter to nationalist ideology, traditional norms and the distribution of power in ethnic networks. But individuals can connect to other contexts without the oversight of their own.
Permalink to Comment5. Zbigniew Lukasiak on November 25, 2003 6:32 PM writes...
I don't know how to formulate it most politely but - I'd like to know the source of this information: "Game theory holds the best strategy is to trust but verify with each iteration."
Permalink to Comment6. Ross Mayfield on November 25, 2003 11:18 PM writes...
Polite enough. With a basic prisoner's delimma the optimal strategy is defection. But if iterative, the optimal strategy is to cooperate until when the other party defects. See Nash Equilibrium.
The trust but verify part is Lenin, but that's probably not your question.
Permalink to Comment7. ZbigniewLukasiak on November 26, 2003 3:37 AM writes...
I meant a link to the actual article. The Nash theory is a general one. The result you mentioned is not - it must be a feature of some models but not all. In fact I think that in zero-cost identity it is not true. I'd like to know the exact constraints of the model.
Permalink to Comment8. Zbigniew Lukasiak on November 26, 2003 4:41 AM writes...
This is still very vague thought but I think the family/institution centric notion can be generalized to something I called global and local reputation. The global reputation is the institution centric social capital, the local is the family centric. But the same categorisation (based on trust defined as some probability measure) can be used for message filtering based on taste - this is the local part, and on trolling/flamebaiting - this is the global (universal law) notion.
I've created a page on my wiki for discussing this idea: http://artcom-studio.com/kwiki/index.cgi?GlobalAndLocalReputation
Permalink to Comment9. John Abbe on November 26, 2003 7:29 PM writes...
Within the currency metaphor, another way of making Ross' last point (that aggregation of social capital is difficult) is to say that social capital comes in myriad local currencies.
Permalink to Comment10. Lucas Fletcher on November 26, 2003 8:43 PM writes...
"This is one reason you can't trade goods or cash for social capital."
How about the obvious reason: social capital is an evaluation of the usefulness of the person herself. You cannot trade social capital any more than you can trade people.
The interesting space is where codified social capital via reputation systems (trust) begins to be used for things other than monetary transactions, namely deference. Entrusting someone to supply you with goods or services is the old economy. Entrusting someone to ease the burden of the complexities of modern life by making decisions for you represents a facet of the new. For example: who to vote for (we don't have time to research the true intentions of every candidate), what to believe (the social capital of large media institutions for example has plummeted in the last few years leaving a vacuum of trust), etc.
Permalink to Comment11. Valdis on December 1, 2003 1:55 AM writes...
> We all have our internal calculators keeping
> tacit track of who is wronging and righting,
> the health of the relationships and adjusting
> our actuarial tables according to experience.
Yes, we do... but these evaluations are not exact like 'calculators' AND they are greatly influenced by the very social network we are tracking! And you may have different rules of thumb for each network you are embedded in. Very fuzzy, messy, and variable...
Permalink to Comment