Failure of Contract Enforcement in Second Life and its Economic Effects

I think it may be impossible to establish any bank as we conventionally think of it in Second Life because there are certain governmental roles essential to banking that Linden Lab has affirmatively declined to provide. It is these failures that will ultimately keep Second Life “just a game” despite all their desires to the contrary. Here I explore one of the most critical of Linden Lab's failures to act as a government through the recent example of financial services in Second Life.

What I call “piggy banks” have existed in Second Life (SL) for years. Third-party merchants such as SLboutique.com or SLExchange.com allow you to put L$ on deposit with them, but as L$ can only be added or removed from the game by the Lindens, these L$ are simply transferred into a “wallet avatar” belonging to the third party merchant site. This merchant “company scrip” can then only be converted into merchandise from the online merchant or back in to L$ by transfer from the wallet avatar to you. The entirety of your L$ on deposit could disappear if the online merchant either chooses to close, refuses to honor their scrip, or is shut down by Linden Lab (LL) for “any reason or no reason” as their Terms of Service allow. I have no reason to believe that those events will happen, but they can, and so far as I’ve seen, that idea doesn’t seem to bother people much.

However, I call these merchant deposit accounts “piggy banks” as they do exactly what a piggy bank does for your deposit: holds it until you decide to take it out.

There has been at least one investment bank that has been in operation for about a year that offers interest bearing deposit accounts. I decline to mention the name of this bank because there have been some accusations that the rates of interest that it pays are too high and therefore must be fronting some illicit activity or that it is some sort of confidence scheme. I’ve not seen compelling evidence that it is or is not. Another such interest yielding investment firm has recently announced its opening, paying even higher rates of interest causing some to make the same allegations of malfeasance. I offer no judgment on the allegations in large part because the privacy shields put in place by the Terms of Service of LL make fundamentally unknowable the veracity of these investment banks and simultaneously difficult for the bankers to establish the validity of their operations. It is worth noting that the first of these investment banks very clearly states:Let's say I want to open the First Reliable Bank of SL. What would I need to do? I could buy a surety bond to cover all deposits. Now the first hurdle pops up: what would this bond be based on? It cannot be the L$ because unlike apples or dollars, they don't exist. True, dollars only exist - like any fiat currency - because people imagine them to and enough people share the illusion that it works. This is not the case for L$ because for only somewhere between 5000 and 50,000 people believe they exist which might be a sufficient threshold except for the fact that LL could press the Big Red Switch tomorrow and all L$ would cease to exist. Moreover, in the event of a closure of SL, there would be no one who would honor the concept of an L$ for anything of any value. So my bond for the FRBSL would have to cover a case where the value of all deposits became nil.

Which brings us to the question of what the value of the deposits would be in the first place. Should it be the current L$ to US$ exchange rate as of the date of deposit? Okay, which one? Unlike dollars to yen - which are valued by an open, efficient market, there are multiple exchanges for L$ which are only loosely correlated in day-to-day rates. So we'll pick an arbitrary one (e.g. IGE) and hope that IGE doesn't decide to stop trading in L$.

So FRBSL will establish a public surety of US$10,000 to cover the initial deposits. Where shall that bond be obtained from? Conventional underwriters like to have some actuarial assessment of risk in order to issue coverage. As noted above, the risks are as large as the deposits in the case where SL shuts down. What is the likelihood of this happening? If I were a conservative underwriter, I'd have to assume a probability close to 1.0 of this happening in the next year. SL is small, and as a recreational endeavor is extremely subject to substitution effects. That is, a new competitor could essentially eviscerate the customer base of LL by releasing a better mousetrap tomorrow. Indeed, whether or not this may soon happen, Linden Lab appears to be acting as if they think it a distinct possibility.

Any real world underwriter worth their name would likely charge $X for a probable total loss of an asset valued at $X within the next year. I am not saying that SL will fold in the next year, just that as a privately held firm of not great size and closed books, a prudent underwriter ought reasonably expect that outcome given the typical survival rate for firms of the size of Linden Lab.

But I could still establish a surety by purchasing a public performance bond equal in value to the L$ on deposit with FRBSL. Now I’ve just moved the fiscal burden of guaranteeing deposits from Linden Lab, not to the underwriter, but to me. Following the same rationale as the underwriter, I’d have to conservatively expect to loose that cash in its entirety. So what would motivate me to do that?

What motivates real world banks to provide services are two things: first, in the US, the federal government - through the Federal Deposit Insurance Corporation (FDIC) – acts as an underwriter to all banks, charging each a small premium for that protection for individual banks and bringing the considerable weight of the US economy to serve as a cushion against a total failure of the entire banking system. Although this makes running a bank less risky for the individual bank, that risk reduction alone is not sufficient to cause banks to want to operate or else they’d be mere “piggy banks” per my definition above.

What causes real banks to want to operate is the prospect of profit. How is this profit obtained? As the old saw goes, by banks following the 3/5/3 rule: pay your depositors 3% interest, charge your borrowers 5% interest, be on the golf course by 3pm. The term “borrower” is new to this discussion. Without the ability to loan money and charge interest on those loans, banks have no way of generating profit and thus no reason to open their doors in the first place. And it is exactly when we get to the “borrowing” part that the hole in the governance of SL makes its grand entrance.

In his seminal 1960 book “The Strategy of Conflict”, economist Thomas Shelling explores some superficially paradoxical situations that upon further analysis are perfectly sensible. The one relevant here is that governments benefit borrowers by allowing banks to foreclose on loans. Huh? How could foreclosure benefit borrowers? It is precisely the ability of the bank to force performance (repayment of the loan) that makes them willing to lend money in the first place. Without this assurance, the bank could just as reasonably set fire to the money instead. Thus without the ability to foreclose, borrowers would find no one willing to lend them the money; therefore the paradox is not actually paradoxical at all. Incidentally, Shelling recently won the 2005 Nobel Peace prize in economics, so at least a few people think he has figured out something of merit.

This brings us back to the establishment of a real bank in Second Life. Linden Lab has repeatedly, affirmatively, and unswervingly declined to enforce contracts in Second Life or to even mediate conflicts between players. Thus there is no such thing as foreclosure or enforcing any player-to-player contract. While this does not strictly prevent parties from making real world contracts to be enforced by real world courts, practicalities of a global player-base combined with questions of jurisdiction, applicable law, and the relatively small real world values of Second Life contracts make it essentially infeasible for any player to even bother with real world contracts.

But you may think: what about trust? Can’t we simply operate on that? My short answer to that utopian ideal is that there is a reason we’ve spent millennia inventing and refining contract law and judiciaries to adjudicate them. Because trust can and does fall apart, even more so in Second Life where your trusted partner can disappear forever by pressing Control-Q, has no reputation to be tarnished by reneging on a contract, and can be back in game under a totally untraceable alias within minutes.

This has caused the single, long lasting investment “bank” in SL to have to promise uncommonly high rates of return on deposits in order to get players to be willing to assume the risk of placing funds on deposit in a world that is based on absolutely nothing more than trust. Trust that involves at least two parties of uncertain prospects: Linden Lab, for the reasons noted above, and the in-game financial services firm for which the same reasons apply but which is significantly smaller than Linden Lab.

Despite all their talk about becoming the next generation internet or the metaverse or whatever lofty dreams Linden Lab has publicly claimed as the potential of Second Life, these dreams will likely not be realized unless Linden Lab begins to assert the costly and time consuming role of government and begin contract enforcement within their world.

The preceding essay is © Malachi Petunia. All rights reserved. It was cross-posted here for purposes of academic discussion. Anyone who says otherwise is itching for a fight.

Please, dear fellow forum denizens. This thread is explicitly not for discussing any particular in-game financial services. Although I cannot stop Ponzi/Not-Ponzi debate from following, I ask that you save those for more relevant threads. Thanks.

I've never trusted contracts to enforce things in the RW, nevermind SL.

I do deals with people because I respect and admire them or people I respect and admire .. respect and admire them.

Do deals with people because they're good, open, transparent, trustworthy people. Not because you think you can write a clever contract.

Of course contracts do have their place.

They're useful to make sure you're on the same page when you're going into a deal. But really.. if the deal stops making sense, both parties will generally renegotiate or something if they're good people.

Contracts are not meant to suck people dry just because of a 'gotcha'.

I whole-heartedly agree; as a mentor of mine put it "once you start pointing to the contract, you've both lost".

That said, models of trust only work where there is accountability and recognition and these, along with enforcement have been affirmatively stripped by LL from their game that they want to be more than a game.

Contract enforcability... oh the WF-Bedazzle-AW brouhaha only serves to highlight that same isuue from another perspective.

You've put your finger precisely on the problem with the various phantom banks in SL. They were started backwards. Banks business is making loans and investments, not paying interest. But most young consumers think only of banks as places to deposit money. Some have taken advantage of that in SL's "ponzi-friendly" system.

Banks existed for hundreds of years without any government regulation or insurance, based on reputation. They did not take deposits. They were owned by wealthy risk-takers who loaned or invested their own money (and that of a few other wealthy allies) with people whose business sense and prospects they trusted, at high rates of interest. And if they didn't pay, they hired warriors to hunt them down and kill them, or have them arrested while in a supportive jurisdiction and take their "pound of flesh."

This was so well known centuries ago that it is the central theme of William Shakespeare's "Merchant of Venice." Shylock didn't have depositors. He had all the money he needed. He needed places to invest his money.

Those who borrowed knew that they had to turn a profit on their business and repay the loans. They knew that the money lenders took steps against defaulters. They also knew that if they repaid their loans, they could get more, bigger loans in the future. The interest paid to the moneylenders provided the profits the moneylenders needed to loan more money to others in the future.

Private bankers such as the Rothschilds funded entire wars for cash-poor kings and queens who needed money, in return for concessions and grants of lands and other things of value. Private bankers did not need to pay interest or advertise for depositors. Only in the recent era did that innovation (and the possibilities for fraud and abuse) emerge, as opportunists sought to expand their banking capital through taking retail deposits.
Contract enforcement is possible in SL, as is foreclosure. It would require far more private structuring of land relationships than most would find efficient. However, with the new regime of setting up 10, 20, and 40 sim "covenanted realms," it becomes more possible. If you've not already, please read the concrete proposal pending in the Political Science forum: "Towards a Theory for Enforcing Contract, Resolving Disputes & Incorporation."
LL's legal and business interests argue directly against it assuming the thankless and profitless (and legally risky) role of being government and contract enforcer. I'd get over expecting Santa to actually come and get to the store to buy the presents and candy for the kids' stockings.

Frank, thank you for the cogent, historical, and insightful reply. I read your Towards a Theory for Enforcing Contract, Resolving Disputes & Incorporation, its commentary and your rebuttal. It is a well considered attempt at solving the tar-baby that LL does not want to pick up.

Although I don't see any theoretical problems with your proposal, I do see one massive impediment to its adoption: it is extremely ponderous - maybe prohibitively so - given the paucity of tools provided by LL.

To my eye, there are much more rapid ways of obtaining the L$ needed to fund an SL venture. Take, for an example, an absurdly costly SL venture, one that would require L$2,500,000 to fund. It seems to me that it would be far easier for the prospective entrepreneur to obtain real world unsecured credit of US$10k and use that to purchase L$ than to instantiate the needed agents and firms within SL. Furthermore, the required lien to secure a loan of that size would require at minimum 10 sims worth of vLand with attendant tier to LL of US$24k/annum. This, of course would net the entrepreneur US$1k of capital at a cost of US$24k which hardly seems like a good return on investment.

So, assuming I got my numbers right (balancing my checkbook is something I'm not real good at) it appears that this system would result in paying LL a fee to continue to abdicate themselves as government.

Kudos for your thoughtful analysis, Introvert.



Ponderous is a fixable drawback in my opinion. It's not the same as having a problem that in principle cannot be solved, or would require prohibitive amounts of resources to solve. Ponderous brings to mind first drafts and early versions of proposals that can be streamlined into something workable.

Several forum posters have pointed out the need for enforcement of contracts in SL, for example here in the Law Society group forum. We differ on whether enforcement should be by LL, as you recommend, or whether enforcement of contracts is best accomplished by new tools in the hands of residents. I don't expect LL to take on contract enforcement, because of the expense in staff hours, the potential risk and the uncertain returns in terms of customer satisfaction.

Your analysis seems to support that view, although some of your wording suggests a desire for LL to be more like a government, complete with judicial branch. I don't want LL to be the government, because it's unlikely that a company could do it well and be profitable. I don't want a fellow resident to be the government either. My support for better tools in the hands of residents is qualified by the assumption that we'd be worse off if government-like tools are available in conjunction with dictatorship group structure for land groups. (This caveat has nothing to do with non-land groups, such as an event group, designer group or club group). So the question becomes one of selecting tools that could be used by any residents for productive economic purposes, not for the founder of a dictatorship structured land group to exercise judicial branch control over the residents on that land.




Your example assumes no risk sharing by a group of investors with a shared interest in the venture being funded. But consider the example of a group of content creators pooling resources to establish a joint venture. It's not the same as just buying land to rent out, a very dubious proposition for main grid land. The group of content creators would be investing in something to leverage their existing income-generating capabilities. This example strays from your original topic of banks in SL, of course. But the broader economic discussion of what works in SL does provide some useful context.

Agreed, contract enforcement might not be a goal that would justify the transaction cost of the theoretical multi-owner sim with covenants. But if a multi-owner sim was created for other reasons that justified the transaction cost, one of the secondary or fringe benefits would be the capacity to offer contract enforcement.

This capacity might be practically limited to relatively minor contract amounts, but could be conducted using the "screen identities" of SL without the "decloaking" necessary to resort to FL tools. For example, if three characters wanted to get together and own and operate a whips-and-chains hard core S&M brothel for transgender furries, but didn't want to reveal that in FL, they were a high school teacher in Kansas, a real-estate saleman in Tampa and a family restaurant waitress in DesMoines, the whole deal might be done and enforceable "in world."

For big development deals involving tens of thousands of US Dollars, such as the rapid purchase, development and operation of a revenue-positive 10-sim destination resort and theme park by a lessee, a FL capitalization and contract (with the disclosure of real names and credit references or security typical to such) would make more sense. It could be enforced using customary FL tools.

Once it was up and running, however, the secondary contracts for vending concessions in the park might be created and enforceable against the lessee's interests in the parcel using SL tools such as that suggested.

I analyzed the problem in response to the theoretical challenge presented by the constant refrain of "contracts cannot be enforced in SL."

Cash up front before work done. Insist on it. Watch them stop hiring people.

Points taken, Traxx and Frank. You are right about the theoretical enforciblilty and my L$2.5M example was - upon reflection - a bit of ad absurdum.

I think we are fundamentally in agreement, just looking at it from different angles. My point was as the sole owner of "force" in SL, LL is most readily suited to handle enforcement and that would be the most efficient route to contract enforcability and the promotion of the economy they elsewhere claim to desire.

However, it is also true that LL doesn't want to be in the enforcement game - which probably means they'd not be very good nor efficient at it. Since that truly seems to be the case, Lardner's proposal is an effective way to handle enforcement with the meager powers provided players. Unfortunately, my gut feeling is that it is too costly to secure the level of contacts where it would be most useful.

In other words: we're still likely stuck with a trust model for most practical purposes which I expect least advances the economic development of a metaverse.

I do think that bit about the "phantom banks" of SL being started backwards deserves to be bronzed.

Thank you, Introvert Petunia, for making a very useful contribution to the discussion and consideration of how to create some of the necessary financial structure in SL.

I particularly like the approach of focusing on the customers, and what financial needs residents actually have, as the foundation for determining how to structure some analog of a banking system.

Obviously the major financial demand would be in loans to allow the purchase of land. Many residents have ideas on how to make use of land, either for profit-making or personal purposes. This is a major topic, which I will return to below.

A second category would be the merchants and designers in SL. One advantage content developers have in SL is that no actual production investment (other than time) is required. You don't need to have access to capital to begin turning out buildings, clothes or other products. There would be a need to access loans if one wished to expand rapidly and open a large number of stores with associated rental expenses. This would be more like a line of credit to cover operating expenses.

The third category (and likely the most minor in terms of value) would be people who wish to engage a content creator, but don't have all the cash in game they need.

I have been looking at this requirement bearing in mind the operation of a credit union, which is many ways is similar to and possibly a first step towards the corporation structure Frank proposed. A credit union pools the resources of members to make loans for various purposes, charging interest which is returned to members in proportion to their investment. Loans are only made to members, who in the purest sense of a credit union have some common affiliation. The membership, through elected representatives, makes decisions on loans to reduce the risk to the group.

The management of risk is another point to consider. Government deposit insurance is based on the premise that most risk is localized - an economic downtown or bank failure is likely to only affect a small portion of the banking community. It can't be of any value in the case of a major national or planet wide disaster (e.g. killer asteroid). Similarly any organization in SL can work to reduce risk of certain types of events, but can't handle the system-wide hit - i.e. LL pulling the plug.

For secured loans, as there is no equivalent of a land title office in which mortgages can be registered and land reclaimed, the credit union would have to work the opposite way - rather like Frank's corporate structure. The credit union would purchase the land and hold it until the mortgage was paid, at which point ownership would be transferred to the mortgagee. During that period, the mortgagee would have full rights to build on the land and contract for rentals. The payments would have to cover tier and loan amounts. Given that the game may change, and players could come and go with little warning, it might be best to restrict the term of such a loan to one year or 18 months - that would be adequate to repay, at a couple hundred dollars a month, a loan in the thousand dollar range plus associated tier.

For unsecured loans, there could be rules to loan certain amounts to members who have met various criteria (length of membership, amount on deposit, etc.). For higher amounts or higher risks, a co-signer process could be used to help guarantee performance.

We have to distinguish between the principles and procedures under which this credit union would operate, and the actual SL mechanisms used to implement those procedures. The limitation on ownership and activity to a single avatar is probably the major impediment. This is an aspect that LL could change without getting into the messy mechanics of being a government.

Ferren, a very useful concept for a credit union.

If I understand it, you're suggesting that a well-capitalized group could acquire one or more sims and form a credit union like you describe to finance the sale of subdivided parcels to those who join the credit union. As borrower members got established and paid off their loans, they could deposit excess cash with the credit union to make more loans secured as you describe. The credit union could pay dividends to those depositor members.

The land held in bulk as security might even generate dwell rewards and save on tiering fees. Those rewards and savings could be shared with depositor members or go to compensate management for their work in running the credit union.

Once all the subdivided parcels were sold and enough of the loans repaid, the credit union should have recovered its capital and be able to buy another sim and repeat the process, retaining the depositor-members that own property in the first sim. Management would be expected to run the credit union on a financially sound basis and to issue monthly financial reports, much like Neualtenburg does.

For those exploring this approach, it might help to explain the loan security concept by comparing it to a "rent-to-own" arrangement often used by folks without credit histories to acquire big ticket consumer goods.

At the risk of sounding a little flip (which I'm not trying to do!) there is an interesting assumption that has been made.

The assumption that contracts are actually enforceable *outside* of SL.



Take the classic 'contractor' problem.

So you want to add a gazebo, or some landscaping, or fix a leaky roof.

You call a contractor, and all the paperwork is filed, licences checked... &c.

So what happens?

Halfway through the job, the contractor disappears for a long time. Turns out, he has another job across town that popped up. Though when you go over there to have words with him - he's not *there* either. He's on a third job...

Now, is it worth taking such cases to court?

Sure, you could. Or you could short-pay and let him try to recover money from *you*.

But at bottom, what is going on is this: contractors that don't constantly scramble for jobs are too expensive, too hard to schedule, and generally go out of business. There is no 'better contractor' out there waiting to do business with you.

Yes, the contractor might not get that last payment from you, and yes he will probably finish eight weeks later than planned. But at the end of the day the system (such as it is) works.

When your roof leaks, no matter who you call, you are dealing with 'the way things work in the industry'. Because consumers voted with their dollars and *made* the industry the way it is.


Point is this: when value is relatively small, trust is far more efficient than the overhead of contract enforcement. Is anyone really going to attempt enforcement of a contract worth, say, less than half a million $L (roughly 1850 USD)? Attorney fees alone would make it a waste of time.

The problem may not lie so much within SL, but the fact that we are still a micropayment society.