Circle of Experts - Brain Flood Blog

Tuesday, January 02, 2007

Research Pricing

I enjoyed participating in AQ Research's recent conference on "The Future of Research", last December 5.

I finally had a chance to edit and post my notes on the panel on Research Pricing.

Speakers:

Lisa Shalett, Chairman and CEO,

Sanford C Bernstein Paul Spillane, CEO, Soleil Securities

Chair: Albert Alonzo, AQ Research Lisa Shalett:

If research is really a discrete model, you could say that should have a fixed price.

But I believe research is actually really an advisory service, an interaction, part of a broader process----part of the mosaic of ideas.

No money manager on the planet will say he has just one source for investing. Research should be priced in direct proportion to the value creation or value destruction.

By comparison, money management industry prices its service in proportion to its own value creation or value destruction. Unbundling and transparency are two separate concepts.

The idea that research pricing should be tied to volume or fixed is another concept.

I believe strongly that research should be tied to volume.

Paul Spillane: For us to be successful, we have to determine a fee. We are the only part of the finance business that provides product for free.

I don’t want us to be paid based on our stock-picking performance. 1-2 fund managers, and a batch of hedge funds, have given us a price schedule: you bring in a company manager, I'll pay X. you provide an expert, I'll pay Y. you set up road show for me, I'll pay Z.

We're trying to ascertain value of what we provide. Our analysts only get paid if customers say, "Pay them". It's all client-directed.

Alonzo: Independents have always been transparent, but sell-side have dragged their feet in complying with menu pricing. Shalett: Trading commissions should be going down slowly.

There's no reason why they should be flat or rising. The wool is being pulled over the money management industry's eyes.

Marginal cost of executing /clearing a trade is close to zero, and Sanford Bernstein is not even a scale player in this business.

If you want to save money on research & trading, start there first! Spillane: If you're head of research at a bulge-bracket bank, you focus your effort on minimizing outliers, so no one can sue you.

You're not focused on creating alpha. Money management industry is afraid to say what they're paying for research. If a money manager says it's paying 2 cents for research someone can sue. But if you say you're spending it on execution, it's harder to complain about it.

Look at Jefferies settlement. Shalett: Broader public policy question: what does research provide?

Buy side says they're paying for liquidity, for there to be a dialogue. Trading cost curve has gone down 6 pct. annually since 1975, to 3.5 cents now.

The regulators want you to be democratic, give access to everyone.

Pricing mechanism should be scale neutral, to not overly advantage the largest money managers. Alonzo: Why do indies still get paid so much less than sell-side?

Spillane: Because we let them! We don't know what we should be charging. We don't have enough senior-level relationships. We don't know if we're being overpaid or underpaid.

You have to be willing to walk away from underpayment. Shalett: We don't put a formal price on our services. But we use the concept of fair share.

We use polls, Greenwich polls. We're paid 1/3 of what we should be paid, if you compare our ranking in polls with the total payments to research providers.

That hasn’t changed in 13 years. The reason is that the sell-side has a lock on being systematically overpaid.

Spillane: Lisa, you have the power to change this. Just shut them off. Shalett: We'd shut them off if Thomson would implement the change to make it easier to shut people off.

(laughter) Questions: What would it take for you to get out of execution?

Shalett: It will only happen if we can be confident that we'll get paid 3x what we're paid now. In Europe, implementing CSAs as follows: 8 bps for execution, 7 bps for research, plus 2 bps for proprietary research when paying sell-sides. These sell-side firms are holding the indies' money for 90 days, paying no interest.

If there were a DTC, a Switzerland, for these commissions, I'd join it and exit trading. But I don't believe in what's being implemented today.

Alonzo: CSA gives firms freedom to use traditional payment mechanism. Is now the right time to push on pricing? Shalett: Depends partly on extent to which funds are generating substantial alpha.

Hedge funds are willing to pay unlimited amount for good research. Spillane: How do you get a return on capital that you can't measure? If someone could provide a utility function for research, big banks would sign up and then focus on prop trading.

We're entering dangerous part of cycle, because we've had good returns for several years.

This is first year in 10-15 yrs when more hedge funds closed than opened.

So it's a tough time to talk about pricing. Shalett: Big banks are generating 30% ROE.

They're good businesses. Mayhew: Will we see a more level approach between pricing of sell-side and indie research?

Spillane: We won't move to unbundling until we're forced to. We won't subject ourselves to the soft dollar pool, which is only 7-15% of giant pool of commissions ($14-$15B in US).

I don’t want to be in the kiddie pool. Barry Hurewitz (Morgan Stanley): We have 500 analysts .

Average buy-side person with a $50M commission pool has 11 buy-side analyst. Lisa might have highest per-hour price on the Street, today.

I don’t see why you can't just price your services explicitly; that's what we do. Shalett responds: From 75 to 2000 (decimalization), trading was a commodity, because market was very concentrated.

Most volume going thru NYSE specialists, whether it came from MS or GS, or it was OTC (fulfilled by market-maker). Post-2000, you had fragmentation of liquidity, which meant differential execution quality.

Trading has become a more differentiated product.

Capital commitment actually mattered. Now, pendulum is swinging back.

Technology is reconcentrating liquidity. Buy-side can trade directly with the market.

Vast majority of executions are commodities, or will be in next 2 years.

So prices should be going down. It doesn't make sense that prices go up.

CSAs favor some players over others. Macroeconomics don't stand up to scrutiny, and industrial behavior is being perverted as a result.

Spillane: Morgan Stanley can't price by the hour, because there are too many constraints.

Customers aren't willing to set a price.

There are way too many internal pressures calling your #1 ranked analyst to come to a meeting, go to a conference, etc., to say simply that he bills at $3000/hour.

General | Public Markets Investing | Securities Research

Author: David Teten

Comments [1] | Trackback

Making Investment Research Pay

Following are my rough notes from AQ Research's recent conference on "The Future of Research", on December 5.

The speakers on this very strong panel were:

Scott Lessing, Chief Operating Officer, Citigroup Investment Research Shubh Saumya and

Jai Sinha, Partners, Booz Allen David Weild IV, CEO,The National Research Exchange

For more information on Booz Allen's views on this subject, see their report, Saving Sell-Side Research.

Lessing: This question [of how to make investment research pay] is less pressing than it used to be 2 years ago.

We're coming from a great year for our clients and research.

There's been a shift towards payment for quality of research Jai Sinha:

News of death of research was greatly exaggerated. Components of value of research include distribution and service, as well as actual content.

Any time commissions are not explicit, there is waste. We don’t need to necessarily force change in the model Weild: Quality is in the eye of the beholder.

What is quality to an actively traded hedge fund isn’t necessarily quality to a longer term account or a public company.

I noticed Monika Schulz with Wall Street Letter in the back.

She recently conducted an interview with Candace Browning, the Director of Research at Merrill Lynch.

Candace, when asked how the research product had changed, responded that it has become more "trading focused." It’s clear that public companies are having a tougher and tougher time getting the support that they need from the largest sell-side firms to reach long-term investors.

These accounts don’t generate enough commission volume to merit much attention yet they are the most important accounts for CEOs to reach.

As a consequence, more firms will need to ensure aftermarket support themselves. Let me share one interesting anecdote: I was talking with the CEO of $350-$400M Australasia-based company.

A bulge bracket firm was taking this CEO on an all-expenses paid roadshow in New York and Boston to see institutional investors. This CEO couldn’t believe his good fortune and asked me “Why?” I asked him to share the list of accounts he was visiting and it was clear that most of these accounts (e.g., Tiger, SAC, Goldman Sachs Principal Strategies) were so large as to likely be prohibited from owning a stock of his market capitalization.

The CEO went on to say that all of his conversations focused on industry trends and not his company – a clear sign that they were mostly wasting his time.

We recently did a project for a $750M VC. They were working with a $500M company going public.

The bank with whom they were talking about going public didn’t have coverage on even one company with a market cap under $1billion.

While I might choose that firm to take me public, I sure need to be careful about who is going to support my “little” company in the aftermarket.

Lessing We cover 3000 stocks. We've been expanding small/mid-cap stocks. Weild: We run one funny analysis. We look at daily volume per analyst per stock.

This analytic shows that if you cover a firm within Dow 30, you're still competing for more volume/analyst than smaller cap stocks.

In fact, the median Dow 30 stock generates about 300,000 shares of volume per analyst per day, which drops to about 90,000 shares per analyst per day for the median stock in the S&P 500.

Compare this to the NASDAQ Composite, which is more actively traded than the NYSE Composite, and the median stock generates only about 30K shares per analyst per day.

This is partly the reason why if you're analyst #30 on IBM, you're still competing for more volume than analyst #2 on a mid-cap stock.

A rule of thumb is that you are generating about $0.01/share/day available to research. The balance is siphoned off by program trading and low cost algorithmic trading.

Thus, despite the fact that accounts reject the need for more "maintenance research" on large cap stocks, the fact is that the stocks are so broadly held and generate so much volume per analyst that the economics are compelling.

Lessing: Volume of trading doesn’t drive our coverage decisions; it's investor interest. In addition, there is very little correlation to an idea that we generate and how we get paid.

We can get paid in IBM trades for a small cap idea.

Weild: Yes, but investor interest tends to be correlated to trading volume.

Plus, your definition of small cap is apt to be higher than ours.

We’re looking at sub $500 million market cap companies and Citi is probably closer to $750 million and larger. Lessing: Agreed. Sinha: Unbundling will allow people to pay differentially.

Lessing: there's room for many different payment models. We're very adaptable in how we get paid. Greenwich Associates surveys say that 20-60% of commissions are allocated to research, and rest is execution (capital commitment, execution services, etc). I suspect that FSA data will be roughly comparable.

We'll hold on to a larger percentage of this pot than competitors, and it's because our quality is higher. Sinha: The more street-dependent buy-side firms will allocate a larger percentage to research.

I see a hollowing out of the middle, of the 35th person providing IBM coverage.

Weild: Citigroup provides a whole host of products and services that the independent firms don’t – equity derivatives, enormous sales trading, algorithmic trading (Citigroup bought Lava Trading not that long ago), securities lending, and Citigroup is the #1 underwriter of equities globally.

Citi has a great portfolio of services that go beyond simply research and that gets reflected in commission payments.

Sinha: There is aligning of consumption with spend Spelling: We have a discussion with client about what resources we can commit to them; what % of payment pool should be sent to them.

Audience Question: What threats can you make?

I had a client say, "Show us what you're giving us. If you don't have that, how can you tell us what you're going to take away?" We have 3000 accounts.

There were 4 contacts at one particular account who were heavy users of our research. They were the people we needed to upsell.

Tiers are binary: you turn it on or off. If you want research to be a profit center, how do you do that?

Saumya: How often is there a mismatch of clients who are dramatically underpaying? One of our clients asks, 'What is the research product?'

Lessing: We don’t talk about resources cost in dollars per hour. We talk about where the client ranks with us, and say we believe you're underpaying.

General | Public Markets Investing | Securities Research

Author: David Teten

Comments [0] | Trackback

Social Commerce: Do you want to do business with your friends?

I've recently talked with a few people about 'social commerce'---the idea that our online business activities will both reflect and in part be driven by our personal social network.

My coauthor Scott Allen recently did a market research study on "Transactional Trust in Social Commerce", which provides some context on this.

For example, many people would prefer to buy a used car from a friend or a friend of a friend, rather than a stranger.

The social context is particularly important, in my experience, when purchasing services as opposed to products. Why?

Because the quality of a service varies wildly depending on the motivation and context of the service-provider.

For example, my wife and I are currently evaluating some contractors to do some renovation for us, and a contractor who lives near us and knows some of our friends socially is less likely to rip us off than someone who is a stranger.

Amazon has a primitive version of this functionality, in that I can see that people who like book A also like book B.

I've also seen quite a few startups who are working on various variations of, 'What is an efficient way to buy stuff from friends as opposed to strangers?"

See for example www.stylehive.com and www.thisnext.com.

This was supposed to be a significant part of Tribe's business model, and some of the local services directories (e.g., Yelp, LinkedIn) are also trying to leverage the fact that you trust your friends' (or friends of friends of friends) recommendations. I see several advantages of socializing commerce, in general:

1. Higher likelihood of truth in advertising.

The friend is less likely to lie about how lemony the used car is, because he knows that interacting with you is a repeated game, not a one-time game.

2. Reduced purchase cost because of fewer intermediaries.

By buying a car from a friend, you don't have to pay a dealer's markup.

3. Reduced costs of identifying the right product.

Friends (or friends of friends) tend to have similar tastes. If my friend is (like me) a city dad with a child, then my friend is also likely to have a car to sell me that suits the needs of me and my family.

4. Helping out your friends/your community.

No one does a business transaction unless he/she derives some benefit. You'd rather that your friend gets the economic benefit of selling a car than a stranger.

So does it make sense to use online networks to make commerce more social?

To evaluate against the criteria I listed:

1. Higher likelihood of truth in advertising.

Possibly also true online. However, online we already have measures of reputation that are not dependent on me knowing someone who knows the person in question: eBay's reputation functionality, Rapleaf, etc.

2. Reduced purchase cost.

I think in most cases this doesn't apply online, and in fact the purchase cost when buying via a social network can be higher because I lose the advantage of a broad seller base competing with one another. In addition, somehow the intermediary (e.g., Amazon, eBay) has to make a markup. If I'm buying something online anyway, then I've got access to a shopping comparison engine which will lower my purchase cost to the bare minimum.

3. Reduced costs of identifying the right product.

This is likely true, but only for a small number of products. For many products, my friends are not necessarily more knowledgeable about the product category than Cnet.com---so I should really just buy what Cnet recommends, not what my friends are buying. The one advantage of buying what my friends recommend is that, if conforming is my goal, this helps me to conform.

If everyone else pays a premium for an iPod or Treo 650 then clearly I must buy one too.

4. Helping out your friends/your community.

To some extent this is also true online.

However, I doubt it's a big motivation for many people. Research from the likes of Forrester, Resource Interactive and Morgan Stanley is also beginning to focus on a new generation of consumers they term the Millennials (aka, “Generation Y”) who range in age from 18-26 years old.

Their buying patterns differ from prior generations. The Millennials exceed the Boomers in size, distrust media and have little to no affinity for brands.

As opposed to earlier generations, they value their peers’ advice and validation: therefore, CNet recommendations are less valuable. Jeff Leventhal, CEO of Spinback, observed, "this generation is the largest consumer group to date and will shift the commerce paradigm."

It's clear that many people like doing business with others in their community, or with others to whom they're interconnected. Think how many offices have an internal email list for people to sell sporting tickets, TVs, etc.

Think of the bulletin boards with things for sale that you'll see in many churches or synagogues. However, if you are already doing commerce online as opposed to face-to-face, I'd argue that in many cases it's irrational to make your commerce decisions dependent on your social ties.

In most cases it's more rational to just buy things via a comparison shopping engine (Shopzilla, Froogle, etc.), particularly when buying a commodity good that could otherwise be found in a few block radius.

The good news for the startups trying to do something in this area: first, many people are irrational and will prefer to buy via a social intermediary, even if they get a worse deal. And second, for certain types of purchases buying via a social intermediary can be more rational, especially when the item is a not a common good, but rather a unique or collectible item.

Used cars are the most obvious example, because there are so many ways in which the seller can deceive you about the true value of the product.

If you're a Pez dispenser collector, fellow members of your community can also turn you on to an impulse purchase which you would not have searched for yourself on Shopzilla, but which you are excited to buy because a trusted peer refers you to it. Feedback welcome.

General | Social Software

Author: David Teten

Comments [0] | Trackback