State of the Economy

by Jessica Holyoke on 29/04/08 at 11:55 am

Lowest profit per capita growth ever in Q1 2008

by Jessica Holyoke

Zee Linden recently posted graphs reflecting the strength of the SL Economy here – and Zee is very happy to report that the SL Economy grew by 15%. However, the story is not that simple. By several other measures, the economy growing slowly if at all. For instance, the Quarterly Profit per Capita (QPC) shows an anemic growth rate of 0.18 for the first quarter of 2008, the lowest on record. This suggests that virtual economists may want to take a very careful look at Zee’s numbers.

Quarterly Profit Per Capita posts lowest growth rate on record

Zee’s first graph tracks User-to-User Transactions in SL It is an interesting graph because we’ve never seen user-to-user transactions before. The economic statistics page uses Resident transactions and counts. The raw data spreadsheet has L$ Supply and user-to-user transactions as a page, but only details the Money Supply on that page. The raw data used to create the first graph is not readily available. More importantly, the graph uses two specific terms, user and resident. In Second Life, a user is the typist, the person behind the screen. One user can control many residents’ user-to-user transactions – which would eliminate the transfer of money between alt accounts from the reckoning.

More importantly, Zee Linden is using user-to-user transactions as an indication of Gross Domestic Product, in spite of the fact that it could be anything but. The outside currency exchanges would be a User to User transaction. HippiePay would be a User to User transaction. Paying rent would be a User to User transaction. Giving your friend money so she can buy new hair is a User to User transaction. All of these actions inflate GDP – in many cases without reflecting production. While the purchase of the hair does reflect production, the transfer of the money between friends for the purchase does not.

Another point Zee made was that the transaction rates were annualized. An annualized rate plots the change in an indicator over the whole year as if the latest monthly or quarterly figure is presumed to persist for the rest of the year. This is calculated by multiplying one month’s change by 12 to produce the annualized rate, or one quarter’s rate by four – so the real numbers were transformed a bit in order before Zee used them on on his graph.

Zee Linden used User hours – and again using an annualized rate, Zee stated that the hours also increased by 15% However, the one problem with those hours is that there is an unknown Bot effect. How many of those hours have no economic value to them?

One way to measure the Bot effect would be to compare the Total Users logged in with the Number of Residents who spent money. While this may also pick up Residents who log in and don’t spend money, either way, it removes residents who don’t participate in the economy For June 2007, the last month on the archive with both sets of numbers available, 305,551 residents spent money in world. As of June 30, 2007, the 30 day log in number was 1,091,017 residents. Dividing the number who spent money by the number logged on suggest that only 28% of the world’s residents participate in the in-world economy. While many arguments can be made about how long a Bot is logged in for or if a Land Bot should not be excluded because it does spend money, at least 28% is a starting point. For now, Bot effect is not a good measure because we only have older statistics and it might not be helpful to see how the hours are being used now.

Positive Monthly Linden Flow bounces after falling off a cliff

Zee Linden also uses my favorite stat, Positive Monthly Linden Flow He states that PMLF grew by 11% from December 07 to March 08 – even though, 50,688 total PMLF residents in December to 57,888 total PMLF residents seem to be 14% growth. Even though PMLF has its limits, at its heart, it does measure the number of residents who earn more than they spend If you fill out $1000L of surveys and you spend $900L, you are counted in PMLF.

LindeX activity is skipped because no one has convinced me that purchasing Lindens translates into taking part in in-world production due to the other purposes for which those L$s might be used. For instance, the US dollar is a reserve currency. Other countries buy and sell goods and services in dollars even though those dollars are not used in the US.

So how does the economy track with the figures I used before? Using raw PMLF over Quarter 1 2008, the total number grew by 4.80% from January to March The average weighted monthly PMLF over the same time frame grew by 6.22%.

Quarterly Profit per Capita (QPC) shows Q1 08 shows a QPC of 0.18, the lowest that it has reached on record. Could this have something to do with falling virtual land prices?

18 Responses to “State of the Economy”

  1. Nice try

    Apr 29th, 2008

    I like how the people who study this all the time are wrong, and a tabloid magazine writer suddenly is an expert in economics. Jessica, stick with taking horrid pictures of really ugly freaks. I do hope I did you justice.

  2. Jessica Holyoke

    Apr 30th, 2008

    Two things early on.

    Bot effect should be 28% of logged on users took part in the in-world economy, not 28% did not take part in in-world economy.

    The other point is that Zee Linden aka John Zdanowski has an MBA, not an economics degree. (He also has a MS in Electrical and Computer Engineering.) He is not an economist. Quite frankly, the big virtual economist, Ted Castronova doesn’t like Second Life, at least to my understanding. Nice Try, if you could point to an economist making any claims or deductions about the Sl economy, please share. (Also, I think you are confusing me with Justine.)

    A company’s blog post is not an independent evaluation of anything.

  3. whisper2u

    Apr 30th, 2008

    Some of these retail people think the poor economy problem is all DiSSENTiON’s fault. Check this link out for Epic Lulz:

  4. Cai Pirinha

    Apr 30th, 2008

    “The other point is that Zee Linden aka John Zdanowski has an MBA, not an economics degree. (He also has a MS in Electrical and Computer Engineering.) He is not an economist.”

    Ouch … this comment was so stupid that it completely disqualifies the author as a serious commentator on anything economic.

  5. janeforyou Barbara

    Apr 30th, 2008

    Any economy go up or down depending on your wallet RL or SL.
    It takes time to turn a economy from down to up.If you got a produckt with qualety at prices that make sense and that moust of us demands it wil be the marked that deside if the economy are up or down. And the marked are “SL users”

    Lets say you run a buiness in SL. You create 12 items you need to know why you create it,, why anyone will buye it and you need the right price and the right sale location
    and you need to find the right marked in SL and the right marketing channal.If you then dont sell any items you did one of the tings i metion wrong. :-)

  6. Jessica Holyoke

    Apr 30th, 2008


    Interesting point. I thought about it and checked the Harvard Business School website and it looks like Zee Linden may have taken a course on Business, Government and the International Economy.

    Usually, I look at an economist as someone different than an accountant. (I think the joke runs that an accountant is good with people.) But maybe Zee having an MBA trumps all of my degrees and gives him the potential to be a better economist than I am. I’m not that arrogant to think that.

    However, it still doesn’t change a corporate announcement into being an independent economic analysis. It doesn’t change tilting charts to try to make things look better. It doesn’t change annualizing the rate of growth in order to make the numbers prettier It doesn’t change using those numbers to try to get more capital or to sell your product. There’s nothing wrong with any of that. but its not an independent look at things.

    Some people might be thinking this is an attack piece on Linden Research, Inc. Its not. In gamer terms, Linden Research is neutral. They can do good, they can do bad. That’s because they are people. When they do good, they should be praised. When they do bad, they should be criticized about that. Its just that different people have different points of view as to what is good and bad.

    In this case, publishing the numbers is a good thing because that’s the only way we can get an estimate as to how the in world economy is doing. But just saying millions of Lindens are being sold does not mean its a good time to open up a shop in SL for you, or its a good way to judge how much demand your rentals might have.

    More importantly, this is published for discussion. So far, I’ve been confused with Justine and said I can’t comment on the economy because I said an MBA was not an economist. (Does one class an economist make?)

    The scientist side of me says this is Peer review. This story will be noticed. So if you have a problem with the numbers or the analysis, say so. If you have a better idea, say so. Don’t rely on a self-interested corporation to tell you that the economy is doing wonderful when it builds their bottom line.

  7. Rip

    Apr 30th, 2008

    All of Linden Labs stats are inflated and they are done so intentionally.

    One area is the loggin count, without all the search engine zombies this count could be as much as 30% to 40% lower than what they claim at any given time.

    If you dont believe this, start visiting sims you see on the big map with large numbers of green dots(people markers).

    A large percentage of the time you will find clustered groups of hidden (usual @ 700 meters) non playing search engine zombies protected by security systems.

    Free account camping is bad enough in sl and should be stopped, but now we have Linden labs condoning TOS violations through the deliberate use of non playing static accounts by land and business owners.

    If LL is not enforcing the TOS to bolster their own sagging stats is this ethical?

    Also is it not a violation of trust between LL and its paying memebers, and could it not then be viewed as a breach of the TOS contract we all signed?

    Linden Labs needs desperately to clean up its act and take care of its customers in a ethical way or face some real problems later.

  8. Victoria Wheeler

    Apr 30th, 2008

    I wonder if LL is inflating their log in frequency figures. I must log on ‘frequently’ because I’ve been unable to teleport between sims since the December 2007 update (I have the latest update at this writing). The ‘Labs only acknowledged my problem publically recently, and their recommended workaround was to log out and log back in to my intended destination. Hmm…

  9. Eric Reuters

    Apr 30th, 2008

    @”Nice try” & @Cai

    Zee seems like a good guy, but it’s worth always remembering as Linden’s CFO Zee has not only a *bias* but an *ethical duty* to spin Linden’s numbers in the most favorable light possible at all times. The more journalists/bloggers/pundits are crunching their own numbers, the more likely the truth is to emerge from the conversation.

    My own take on Jessica’s PMLF growth rate chart is that the gambling ban was an SL-economy killer. Think about it — its not only the casino owners, from the big guys to the once-ubiquitous slot machines, but all the avatars in Second Life for no reason other than to gamble all buying clothes, hair, etc etc. The data shows SL is just beginning to recover, and like any maturing market, it’s harder to make a buck than it used to be.

  10. Montana Corleone

    Apr 30th, 2008

    They chose the only rising numbers. Active users are falling, and have been for several months from a peak of 560K odd to 526K last month. They have fallen by about a third in Europe (except the UK) over the last year, and in Japan, and by nearly a half in Brazil. For two months running US active users fell, the first time that has ever happened, and Premiums have fallen for four consecutive months.

    Those are from their own Key Metrics, which is the reason why the February ones were 5 weeks late, and slipped out under the carpet without a blog comment.

    It gets worse for premiums though. And there are under 90K of those, less than 20% of users. From the Economic Stats web page in Support, the last section, Sources and Sinks, you can see in March they paid out L$9m in referral bonuses. This is L$2K paid out for new premiums, paid in three monthly instalments of 500, 500 & 1,000. Thus in March at least 9,000 new Premiums were on board, and as many as 18,000 ie between 10% and 20%. Since numbers went down, and these new ones come in, it follows that more than that were old premiums who either left or downgraded. That’s around an average 15% churn rate every month!

    We know they tidied up the web site, took off the Last 60 Day Logins for instance, when they were plummeting. Now we have the new Communications Manager Katt Linden trying to convince us they are being more communicative by shifting the status reports to a different blog, which is what we now see on the login page and not recent blog posts. And of course, coincidentally at the very same time, it means all those nsty out of service blog posts are magically cleaned up off the SL web site front page.

    Talk about slimy corporate tactics. Still, that gives no excuse for that DiSSENTiON idiot to carry out federal felony offence DDoS and spamming attacks. The FBI is on his case though.

  11. Obscure Doodad

    Apr 30th, 2008

    1) No, it’s not independent data, but a Harvard MBA can probably find some other job if he wakes up one morning and finds he has one that requires him to explicitly lie to 12 million people in intricate detail each month. Salesmen exaggerate, or choose not to mention things, or maybe even emphasize or phrase something differently, but usually people don’t like to have a job that requires them to overtly lie to customers who are buying a product or service from them. If they can get a different job, they will. So these numbers are not worthless.

    2) PMLF excludes land transactions. That exclusion also encompasses payments to LL. The quote from the stats page is “BEFORE Linden Lab Charges are applied to the account”. That means these positive monthly flow people may not be positive if they are paying tier to LL. That always seemed weak to me because while the land cost might be “investment”, the tier is nothing but “property tax”, and that appears in any firm’s operating costs.

    3) But regardless of the stats not including LL charges, the worst indicator of all is the number is 58K residents had PMLF. But there are 89K Premiums receiving weekly stipends. Maybe they have land, but its cost is NOT in the PMLF number. That means that not even the number of residents who are receiving weekly stipend manage to make a profit in the economy. A Premium resident who owns land, pays tier (if he owns > 512) and collects a weekly stipend and sits on his land and builds for fun will be a PMLF count. A Premium who spends all the stipend and more won’t be. So what this is saying is that not even the number of residents who get a weekly stipend of income manage to make a profit in SL.

    2) What is is also saying is that if LL were to apply LL charges to that PMLF figure, it is reasonable to extrapolate a complete crash in that PMLF number. We can even estimate the effect. Category 1 of the PMLF chart is 0-10 USD. All the other categories do NOT add up to the number of PMLFs who are profitable at less than 10 USD/month. A weighted average of the categories yields a number of about $83 USD (excluding the top open ended category), but this is not the measurement we need. What we seek is an estimate of how much profit the “typical” PMLF makes. With the lower category so heavily weighted that number is swamped out. With 32,000 of the 58,000 being under $10 USD (which is tier for only a 1024 parcel), even with some of them owning no land at all, the likelihood is that including tier (LL fees) would collapse the PMLF number.

    3) Conclusion: That 58,000 number is fewer than the Premium account total who essentially get money “free” each week, and it’s only 0.5% of the 12 million total resident count. It is further very likely to become miniscule with properly included LL fee accounting. It looks likely, even with only rough estimates, that the PMLF number would reduce by well in excess of 50% if tier was included. That would be < 29,000 profitable residents — out of the alleged 12 million.

  12. Lord Kamina

    Apr 30th, 2008

    lol Frenchfag.

    Cry moar.


  13. Nidol Slazar

    Apr 30th, 2008

    “The FBI is on his case though.”

    Yeah they’ll be busting down his doors any minute now and confiscating his computer for evidence… LOL

  14. RoFLKOPTr

    Apr 30th, 2008


    “Linden Labs needs desperately to clean up its act and take care of its customers in a ethical way or face some real problems later.”

    Like losing their paying customers, right? O wait… none of their paying customers know how to do anything outside of Second Life, so I don’t think that’ll be much of an issue…

  15. Lance LeFay

    Apr 30th, 2008

    What you fail to take into account is that GDP accounts for much more than your limited view of “production”. If that wasn’t true, the U.S. would not be the rich nation the GDP says it is- the vast majority of our economy is service.

  16. MachineCode

    May 1st, 2008

    Have you forgotten that you spacemonies are worth exactly zero in any real currency? The idea of an exchange rate is an illusion and some unforseen variable could collapse the whole scheme at any given time. And Linden Lab employees will have the exclusively held right not to give a rat’s ass.

    Of course, it’s much less likely that it’ll come apart at the seams than it will just implode due to various things making the inflation absurdly high while interest in an unstable platform falls.

    It’s like global warming vs. a lazer or something. Wear a cup.

  17. rip

    May 1st, 2008

    RoFLKOPTr yes like losing paying customers not search engine zombies, griefers,campers, money launders and all the rest of the freeloaders!

    Opensim and other VW projects very similar to SL or slowly coming up behind Linden and will with luck hamstring or kill SL by the end of the year when they all go Beta.

    Nothing living is ever static and thats even more true of technology.

    Linden will lose their lead, and to be honest, this paying customer will be one of the first to abandon a sinking ship when the time comes.

  18. Heather Heike

    May 8th, 2008

    This discussion has gone rather far afield from Jessica’s original thesis. That being said, let me weigh in on the ephemeral nature of a virtual economy. Those of us with a substantial monetary presence in 2l should be prepared to lose whatever it is we have “on the table,” if suddenly confidence is lost in the LL sim and everyone bugs out. Whether it is US paper currency or $Lindens, the underlying economic systems depend on “confidence in the system.” One is reminded of “El Mystico and Janet” in Monty Python–putting up blocks of flats by hypnotism, a fine thing as long as all residents believe in them. So, in the meantime, we make (or lose) money and try to cover capital expenditures and investment in (un)real property, fixtures, and possibly even incorporeal heridiments–because if the balloon ever goes up, we all end up with a lot of hyper-inflated tulip bulbs to sell and no buyers. Its a game. Not all that different than the life game.

Leave a Reply