by Bob Hiler

For two years during the Internet Bubble, I worked as a Wall Street analyst in Frank Quattrone's CSFB Technology Group covering Internet stocks.

When the bubble burst, research analysts like me got a lot of heat for playing such a central role in the Internet Bubble. I didn't mind the heat--but for a long time, I didn't understand how the bubble had worked... and that really bothered me.

In the aftermath of the bubble, it's easy to point to empty income statements and weak balance sheets. But in the early days of the bubble, these companies blew past every single earnings estimate that anyone put out there. Even skeptics watched in wonder as these companies grew to generate millions and then billions in revenue.

Where did all the money come from? I struggled with this question for a long time, finally writing out the criteria for the ideal answer:

• The number of investors and money invested must increase exponentially over time.
• Early investors must get wealthy, while later investors get screwed.
• The market must get saturated, and then collapse.

And then it hit me: the Internet Bubble was a pyramid scheme.

PYRAMID SCHEME 101

I got my first lesson on pyramid schemes in high school.  Leafing through the local paper one day, an advertisement caught my eye:

Are you a hard worker?  Want to make money working from home?  Call Susan at 703 XXX XXXX to set up an interview.

After years of boring temp work, this sounded like it was worth taking an afternoon off to find out more.

When I got to the company for my interview, I found out that there were lots of other interviewees. In fact, while we waited for the boss to arrive ("from a big regional sales meeting!"), we all crammed into a conference room to watch a video on the company. It wasn't clear what they did, but the video made clear that everyone made a lot of money doing it.

At this point I was pretty curious, so I stuck around until the boss got back. After hearing him talk for half an hour about how much money he had made working for this company, I finally raised my hand.

"I'm not exactly sure what exactly it is that you sell," I said. "Could you go over that part again?"

He smiled as if he'd been waiting for that question his entire life. "Opportunity, m'boy... we sell Opportunity!"  As I looked for the exit, he explained how his company worked.

Apparently, I was supposed to go around selling kitchen knives to all of my mom's friends. Meanwhile, I'd also sell my friends on the "opportunity" to become a distributor, so they could sell knives to their mom's friends.  In return for becoming incredibly unpopular with my friends, I would get a commission if my friends bought a Distributor's Kit with knives they could sell on their own. If they sold any knives, I'd get a piece of that too.

I declined the opportunity to build my own little kitchen knives empire.  But when I got home, I started reading up on pyramid schemes.

WHY PYRAMID SCHEMES ARE DOOMED

As I quickly learned, pyramid schemes are a lazy man's dream. As long as I could recruit my friends to become Distributors for the low, low price of $100, I wouldn't actually have to sell any knives. And if my Distributors sold their friends on the same sucker's deal, I could retire and cash checks without actually selling any knives myself. The boss was right: he was selling Opportunity, not kitchen knives.

This is the key to every pyramid scheme scam: selling Opportunity to new recruits. As long as each level of the pyramid recruits others to the pyramid, everyone generates astonishing returns. But the pyramid's growth sows the seeds of its own doom. If everyone recruits ten people, the pyramid grows at an exponential rate, with each level of the pyramid ten times larger than the one above it.

For our pyramid of kitchen knive Distributors to hit eight levels, we'd have to convince a billion people to buy Distributorships.  If we failed (and no one actually bought the knives), then the pyramid would collapse. If you got in early, you probably had plenty of time to cash out. If not, you had some decent kitchen knives, but lost all your money.

HOW DID THE INTERNET BUBBLE WORK AS A PYRAMID?

So was the Internet Bubble a pyramid scheme?

As I've learned about pyramid schemes, I've noticed a few patterns.  No matter what the scam, there are four consistent themes to how they worked. 

Let's apply these four themes to both kitchen knives and to the digital media stocks that I covered.

1. Pyramids Tout A Revolutionary High-Return Strategy.
   
   As I learned from the pyramid boss, you could make a lot of money selling kitchen knives.  In fact, the parking lot was filled with the cars that early Distributors had bought: mostly Cadillacs and Thunderbirds (remember, this was the 80s).  Promising a high school kid enough money to buy a Thunderbird is a pretty compelling high-return strategy.
   
   The Internet Bubble got its start when Netscape went public with an explosive IPO.  After those first-round investors got a taste of those high returns, more and more investors became desperate to get their money into Internet companies.
   
2. Pyramids Have New Investors Transfer Money to Old Investors.
   
   In the kitchen knives pyramid, my friends were supposed to buy Distributorships from me.  In other words, the new investors in the pyramid gave the old investors money.
   
   In the Internet Bubble, the money trail also flowed from new to old. Following the money requires a bit more detective work, though. Unlike in a traditional pyramid scheme, new investors didn't just hand their money to old investors. Instead, they followed a three-step process:
   
      1. The high returns enjoyed by old digital media investors (think Yahoo) encouraged new investors to invest in new digital media companies (think TheGlobe.com)
      2. To get traffic, these freshly funded digital media companies (TheGlobe) bought ads from more established digital media companies (Yahoo)
      3. The stocks of the old companies (Yahoo) went up, making the old investors money.
   
   This established an indirect link that transferred money from new investors to old investors.  But don't worry for the new investors (TheGlobe).  With Yahoo stock soaring, tons of money poured into the digital media sector, funding new companies to buy TheGlobe's ads.
   
3. All "Levels" Of The Pyramid Sell The Same Stuff To Each Other.
   
   Most pyramids require all the participants to sell the same product or service.  Our friends the kitchen knive sellers are a good example: we were all supposed to sell kitchen knives to our mom's friends and to each other.
   
   Digital media during the Internet Bubble worked the same way: everyone sold each other online ad banners. With digital media companies largely buying ads from each other, online ad rates surged, leaving cash-flush digital media companies as the only buyers. Other customers--especially offline consumer product companies--were priced out of the market.
   
   This resulted in a first mover like Yahoo generating 47% of its ad spending from "pure-play Internet companies" as recently as the second quarter of 2000. Indeed, throughout the Bubble, many digital media companies engaged in "barter" advertising exchanges in which two companies simultaneously bought and sold an equal amout of online ads to each other. For example, even though no cash exchanged hands, 20% of iVillage's 1998 revenues came from such barter transactions. All together, these factors enabled the pyramid-like transfer of money from new digital media companies to older digital media companies.
   
4. Pyramids Make Competition a Good Thing.
   
   I noticed a funny thing about the kitchen knife pyramid. Ordinarily, competition is a bad thing.  But if you're selling kitchen knives, you want more "competition". That's because this "competition" means more customers, each one buying their first kitchen knife kits from their Distributors.
   
   As we saw earlier, eventually these pyramids run out of hybrid competitor/customer/investors.  After a while, there's not enough "competition" to fuel the sales that Distributors need to keep the pyramid going. And once the pyramid runs out of Distributors to recruit, the pyramid starts to be awash in excess inventory.
   
   Why is that? Well, with no new money flowing into the system, everyone in the pyramid stops buying. And if nobody outside the pyramid is buying, a huge inventory glut develops--with nobody demanding the product. Thus, saturation reveals massive overcapacity which destroys the economics of the market.
   
   Substitute "kitchen knives" with "banner ads", and we saw the exact same dynamic in the digital media market.
   
   The more websites popped up, the more revenue websites generated by selling online ads to each other. The spectacular growth in high-margin ad revenue attracted more companies and money to the digital media space, until the number of new ad-buying websites needed to sustain existing sites became impossibly large. (Even VC's couldn't invest in digital media companies fast enough to keep up.) Once this happened, saturation occured: a huge ad inventory glut developed and the suddenly revealed overcapacity destroyed online ad rates.

HOW DID THE INTERNET BUBBLE GO UNRECOGNIZED?

So, Internet Bubble = pyramid scheme.  How the heck did we miss that?

In comparing kitchen knives and digital media, one big difference keeps coming up. The kitchen knife pyramid scheme was carefully designed and built by a centralized founder. In a lot of ways, the digital media pyramid scheme was the world's first decentralized pyramid scheme.

Take a look at the way that money flowed through the decentralized pyramid scheme:

Compare this to the historical treatment of pyramid schemes. The SEC and FTC has cracked down on undisguised pyramid schemes for decades, forcing them to the society's margins. The failure on the part of Analysts like myself to pierce the decentralization of the digital media pyramid scheme gave the Internet Bubble the validation - and access to IPO millions - that it needed to suceed.

CAN WE IDENTIFY AND STOP PYRAMID SCHEMES?

Not only is the answer "Yes", but all the heavy lifting has already been done. After all, centralized pyramids have been subject to government regulation for years. Now that we know that decentralized pyramids can happen, we can apply the government's traditional pyramid tests--the FTC's so-called "Amway Safeguards"--to identify and stop illegitimate pyramids.

If a pyramid passes these Amway Safeguards(as Amway, Avon, and Tupperware have all done), then they are highly unlikely to collapse. Rather than tar these companies with the "pyramid scheme" brush, we call them Multi-Level Marketing (MLM), Network Marketing, or Direct Sales companies.

So what differentiates illegal pyramid schemes from legitimate MLM companies?

All pyramids must collapse if the major source of cash comes from the pyramid's participants, instead of selling a good or service. As long as I'm not selling kitchen knives, I'd better be recruiting new members to keep the pyramid going.

However, a pyramid is not doomed to collapse if there's real money coming in.  In other words, if my mom's friends are actually buying kitchen knives, maybe this business won't collapse after all.  (To be fair, I'm pretty sure this was the plan of the kitchen lnife company whose recruitment session I attended).

Building on this logic, in 1979, U.S. courts established three "Amway Safeguards" to differentiate between genuine MLM companies and illegal pyramid schemes.  By applying these tests to the Internet Bubble, we can see whether or not existing laws would have protected Internet investors:

1. The 10 Customer Rule
   
   Members of legitimate MLM operations must sell to at least 10 distinct customers outside of the pyramid. In other words, at least 10 of Mom's friends need to buy the knives. This prevents a pyramid from generating revenue solely from new recruits, ensuring the pyramid's eventual collapse.
   
   The intent of this rule is to encourage retail sales as the primary means of supporting the pyramid.  So did digital media companies pass this test?  
   
   Unfortunately, they failed. While some digital media companies may have sold to more than 10 non-Internet customers, the vast majority of digital media companies sold primarily to other Internet companies--and did so successfully for quarter after quarter.
   
   Most digital media companies failed to find other revenue streams. In other words, they never got around to selling kitchen knives. With dismal content subscriptions and low e-commerce volume, the vast majority of Internet Bubble companies don't satisfy the first Amway Safeguard.
   
2. The 70% Rule
   
   To keep the pyramid focused on selling knives and not Distributorships, members have to sell 70% of the items they've bought before they can place another order.
   
   It's tricky to apply this test to digital media companies. After all, how do you sell 70% of the ads you've bought? An analogous test would be whether or not digital media companies either profitably sold--or helped somebody profitably sell--a real product or service to at least 70% of the customers they got from that ad purchase.
   
   Once again, digital media companies fail the test. Users typically clicked on less than 1% of ads--and of those who clicked, few purchased something. Even if a site had a great conversion rate of 5%, it would take 5,000 viewers of an online ad to sell a single product. With online ad rates soaring, buying customers became a money losing proposition.
   
   Strike two.
   
3. The Buyback Rule
   
   A legal MLM company has to repurchase unsold and unopened inventory from Distributors. This prevents knife sellers from recruiting new members just to make enough money to cover the knives they had to buy to join the pyramid.
   
   Dr. Koop and eToys both bought huge amounts of ad inventory from digital media portals. They were unable to turn that ad inventory into retail sales. So when they went bankrupt, did the portals repurchase their unused ad inventory?
   
   Once you've stopped laughing, you'll probably agree with me when I say that Internet Bubble companies fail this test as well.

Existing regulations would have served their purpose for the Internet Bubble. The problem lay not in the regulations, but in actually identifying the Internet Bubble as a decentralized pyramid scheme.

WHAT'S THE POINT?

Identifying the Internet Bubble as a decentralized pyramid scheme after the fact may seem like 20/20 hindsight.  But I think it's enormously important to understand that many bubbles are formed by decentralized pyramid schemes.

Actually, I've been using "Internet Bubble" and "Digital Media Bubble" interchangeably, but I should probably be more precise.  The Internet Bubble was made up of several related bubbles: the Digital Media Bubble, the IT Services Bubble and the Internet Incubator Bubble come to mind.  And not all Internet companies were Pyramid Schemes - Amazon, Google, eBay, and Overture are all healthy companies with positive cash flow.

In future articles, I'll take a closer look at these other bubbles, as well as mapping out this generalized theory of Bubble Investing, and how it tells us:

• When bubbles peak, and how to avoid getting burned,
• A checklist to identify decentralized pyramid schemes and the bubbles they cause, using the telecom sector as a case study,
• How non-bubble companies can be affected by bubbles, and
• Why the future will see more frequent and violent bubbles.

If history is any indication, bubbles ain't going away. They've plagued us since the the Dutch tulip mania in the 17th century and the South Sea Bubble in the 18th.

But you don't need any fancy analysis to avoid pyramid schemes yourself.  Don't get distracted by flashy technology or the lack of a centralized ringleader.  Just ask yourself: are these guys selling Opportunity or kitchen knives?