With Second Life concurrency trending downwards, people are looking harder at the Second Life economy and the Linden Dollar exchange, with some speculating that there is a fairly heavy thumb on the exchange-rate scales at the moment.

Before someone brings up the topic of fiat currencies, I’ll do that myself. Yes, the Linden Dollar is a fiat currency, however every national currency in existence today is presently a fiat currency.

The US Dollar itself became a fiat currency in 1973, during a process called ‘Nixon Shock’, when US President Nixon decoupled the US Dollar from material resource. Despite the name, Nixon Shock was a well-received and popular move at the time, and has insulated the economy from subsequent major stock-market crashes.

Nevertheless, the status of national currencies as fiat currencies causes some issues as well.

While a freely floating national money has advantages, however, it also has risks. For one thing, it can create uncertainties for international traders and investors. Over the past five years, the dollar has been worth as much as 120 yen and as little as 80. The costs of this volatility are hard to measure (partly because sophisticated financial markets allow businesses to hedge much of that risk), but they must be significant. Furthermore, a system that leaves monetary managers free to do good also leaves them free to be irresponsible—and, in some countries, they have been quick to take the opportunity. – Paul Krugman,  “The Gold Bug Variations”

The basic principle of a modern currency is to ensure that (for the most part) the amount of currency in circulation is approximately equivalent to the value of goods and services within the nation – a form of supply-and-demand equation where that value is the demand, and new currency is supplied to match it. As long as this is true, the economy (and its currency exchange-rates and prices) remain relatively stable. The state, therefore, attempts to determine that value, and creates or destroys currency as indicated.

Creation of new currency is more common, as the combined value of goods and services tends to grow over time, however destruction of currency to rebalance slumps in that value are not unknown. Governments also create new currency to cover deficits – which leads to inflation (the rise in prices charged for goods/services, wages paid to employees and so forth) as the currency supply starts to notably exceed the combined value of goods and services.

Okay, so we’re up to speed on all of that? Now, back to the Linden Dollar.

The Linden Dollar is superficially similar to other fiat currencies, but differs in several important respects.

Linden Lab is the issuer, and creates and destroys Linden Dollars. Users buy Linden Dollars used (from other users of Second Life) or new (from Linden Lab). There are no visible distinctions between existing Linden Dollars and new ones. The user makes a purchase, and receives Linden Dollars without knowing whether they are from another user, or freshly minted by Linden Lab.

On the other hand, a user who is selling Linden Dollars, is always selling to another user. Linden Lab never purchases Linden Dollars from a user.

Like any currency-issuer, Linden Lab creates Linden Dollars according to a determination of the total value of goods and services. The Lab uses a different mechanism to gauge that value and estimate the demand for new Linden Dollars, however.

If a user wishes to purchase a certain amount of Linden Dollars, and that purchase cannot be fully satisfied by user-supplied Linden Dollars being sold on the exchange, Linden Lab creates the difference (using a special account called Supply Linden), and pockets that portion of the transaction. It’s at once both profitable for Linden Lab, and does a surprisingly good job of insulating the economy and the exchange rate from all sorts of lumps and bumps. Certainly it seems far more accurate than trying to calculate national/domestic products for various nations.

The Lab can get away with such a simple and elegant mechanism because of the high level of foreign exchange involved with the Linden Dollar. The pocket economy is mostly isolated, leading to surprising simplicities like this.

There are also stipends (weekly payments) for premium accounts which are minted as required, but as long as Second Life’s internal economic growth routinely exceeds the total stipend created, it does not provide any unbalancing effect on the Second Life economy.

Conversely, Linden Dollars are deleted from the economy through a number of fixed avenues such as search-listing fees, classified advertisements, content uploads and so forth. These currency sinks need to be neither too large nor too small in a healthy economy, as either would cause a loss of economic stability.

Now, how do the changing circumstances affect the Second Life economy, the Linden Dollar currency pool, the market rates, and so forth? Well, that’s rather more difficult to determine. We have very little data about the state of the Second Life economy, and the quarterly economic metrics posts from Linden Lab actually deliver little or no useful data in determining the actual state of the Second Life economy.

However some time back, I had the opportunity to talk with former Linden Lab CFO (from 2006-2009 and later CFO of Avatar Reality inc.), John Zdanowski, and he gave me a very interesting thing to think about:

Users consistently spend $1 per hour in user to user transactions in second life.  The lindex users spend about $0.20 per hour and Linden Lab’s revenue was also a consistent amount per hour (when I was there).

It’s the most important metric because users are willing to spend a certain amount per hour for entertainment…no matter how you charge them.  That’s the surprising thing.  Changing pricing doesn’t seem affect how much users are willing to spend per hour.

Now there’s an insight you can take to the bank… so to speak. So, those figures were relatively consistent over the course of three years. That suggests that even if Linden Dollar exchange rates start to fluctuate, the spending habits of users will remain relatively constant, keeping revenues for Linden Lab and for users selling Linden Dollars on the exchange, relatively stable and providing an additional stabilising effect on that sector of the economy, which should – by all rights – flow through to the rest of Second Life’s economy.

Unless I’m making an illogical step now, that’s assuming that two key variables remain relatively steady – the number of users and the number of hours.

Now, in a subscription-based economy – common to MMOGs like World of Warcraft –  once again, you have the same essential constant revenue-per-hour-per-user. In a virtual currency scenario like Second Life, more committed users spend more, while less committed users spend less, but it all balances out to consistent per-hour averages.

I’ve got some information sourced from another virtual world (whose name I can’t mention, I’m afraid) which suggests that the average per-hour-per-user principle works there just as consistently as it does in Second Life; and despite vast differences in the number of users, median concurrency and user-hours, works out surprisingly similarly to the figures that Zdanowski quoted for Second Life users.

So, what do we learn from all of this? Firstly, it seems that increasing or decreasing opportunities (or requirements) for users to spend money with an MMOG or virtual world service have very little actual impact on the revenue per-user-per-hour. The users themselves will balance that equation out for you, declining to spend outside of their comfort zone.

Secondly, it would seem that the most important factors for absolute revenues for Second Life merchants and for Linden lab are the number of users and the number of hours (which combined are roughly represented by median concurrency – however the Second Life Marketplace throws a spanner into those calculations by substituting trackable in-world user-hours for untrackable out-world user hours without necessarily reducing spending).

Thirdly, the decline of hours and users has few short-term effects on the stability of the Second Life economy as a whole. Yes, it cuts into profits (for the Lab and for Second Life merchants), and that can really hurt if you’re one of affected. It can cause store and site closures, but things can decline a whole heck of a lot before the economy reaches a tipping point.

That tipping point would be mainland tier and estate fees. These represent large, fixed costs with a limited ability to scale upwards or downwards.

Were a decline in grid-wide absolute revenues to cross below that line (or the fees to rise above it), even for just a few weeks, the Second Life economy would suddenly go from stable and healthy to collapse or near-collapse rapidly (along with the basic user-experience), and with few or no advance warning signs; because we’ve already established that users – be they consumers, merchants or land-holders – won’t spend outside their comfort zones. That sort of situation is one that everyone involved would prefer to avoid.

Tags: , , , , , , , , , , , , , , , , , , ,

Got a news tip or a press-release? Send it to [email protected].
Read previous post: