Generally, the pace of change in the secondary markets for stocks
and bonds has been slow, torturously slow. A tortoise appears
a speed demon by comparison. After all, those profiting from
existing arrangements quite rationally resist change, and they
frequently have the wherewithal to maintain the status quo.
Change does occur. At the end of World War II, most equity
investors were individuals, some of great means, but still individuals.
Gradually, institutions made their presence known and by the
mid-sixties, became a force with considerable clout. The fixed-
and high-commission rates of those years and the associated
abuses generated outcries from the institutional investors themselves,
as well as the powers in Washington: the SEC, Congress, and
the Department of Justice.
This frustration ultimately resolved itself in the 1975 Amendments
to the Securities Exchange Act of 1934. These amendments not
only abolished fixed-commission rates, but set in motion a whole
series of events that led to a significant increase in competition
in the trading of those securities listed on registered exchanges.
The registered exchange markets became linked through the consolidated
reporting of trades (fully operational in 1976), the consolidated
quote system (1978), and the intermarket trading system (1983).
Since these major changes, not much changed in the trading
of U.S. equities until 1997 with the SEC issuance of the order
handling rules. Like the 1975 Amendments, many in the industry
resisted the changes that these new rules ultimately required.
The catalyst for these new rules was perhaps the study by Christie
and Schultz, which concluded that the spreads in Nasdaq stocks
were high and attributed these high spreads to collusion.
Whatever one thinks of this study, the study did galvanize
the frustrations of virtually the same cast of characters that
were instrumental in passing the 1975 Amendments. There was
the usual chorus of civil lawsuits and investigations by both
the Department of Justice and the Securities and Exchange Commission.
Ultimately, Nasdaq was forced to change its governance, making
it more responsive to the needs of investors. The SEC then imposed
the new order handling rules.
Today, dealer profits are down, and it is likely that over
the next several years, we shall see changes of the same magnitude
similar to those that followed the 1975 Amendments. The speakers
today will be addressing these changes.
One thing is for sure: Nasdaq is not standing still. The magnitude
of the changes that have happened at Nasdaq and that will continue
to happen have to be understood in the context of the investor-oriented
mission of Frank Zarb and the NASD leadership team. Our quest
for continual improvement can only be fully understood in the
context of the input and support of what we call our NASD and
The Baruch College event brings together leaders from some
of our key publics. Key representatives from the sell-side,
the buy-side, day traders, the SEC, and academia are participating
in this conference. Many have played a key role in documenting,
critiquing, helping to set policy and, of course, trading on
the Nasdaq market. The diversity of viewpoints represented makes
it impossible for us to agree with each other on each specific
issue; but I think we all share the goal of Nasdaq being a strong
and successful stock market.
It’s worth describing briefly a couple of the key frameworks
within which we at the NASD and Nasdaq consider policy. One
is our value chain and another is the web framework for our
market structure. Then I’d like to discuss some of the changes
that have occurred in Nasdaq in 1997 and some that are planned
for the future.
Each arrow in our value chain points ultimately toward investors,
both retail and institutional— knowing of course that institutional
investors represent millions of individual investors through
their mutual funds and pension funds. Supporting investors directly
are investor interfaces that supply electronic or personal trading
and account management services. Those functions are, in turn,
supported by order entry firms, which aggregate orders and channel
them to the market.
Further upstream are entities that provide liquidity services.
Historically, this has meant Market Makers, but more and more
it also means investors, showing up in another spot in the value
chain as we face up to the reality that, natural investors are
the ultimate source of liquidity.
Furthest upstream are the issuers that list securities on our
market. They account for the vast majority (but due to third
market trading not all) of the grist for the Nasdaq value chain,
which ultimately serves investors.
The disaggregation of market functions described here is somewhat
arbitrary. Much finer disaggregation is possible, for example,
by accounting for distinctions among introducing brokers, clearing
brokers, and correspondent brokers. Chunkier aggregation can
also make sense. Merrill Lynch, for example, is involved at
the investor interface level, as an order aggregator, and as
a liquidity provider. Where is Nasdaq in the value chain? Two
main places. First, it sits in the middle of the value chain
as an information provider and as a transaction service provider
supporting the entry and execution of orders through SelectNet,
SOES and ACT.(5)
Most importantly, though, it is the whole environment—certainly
the massive computer network and systems that make it all possible,
but also the entity that brings together each of the key market
players (including the SEC and NASD Regulation). Nasdaq’s revenue
from its place in the value chain comes from three roughly equal
- transaction service provision through SelectNet, SOES,
- listing fees; and
- information provision (selling the trade and quote data
that are a by-product of the trading process).
The value chain analysis helps us focus on our primary intent,
which is to be the market of choice for investors. Doing that
right coincides with being the market of choice for issuers.
A correlative framework is what we call the Nasdaq web, which
is conceptually related to, but is not the same as the World
Wide Web. In fact, the Nasdaq network that first linked market
participants in 1971 was one of the precursors of the World
Wide Web we know today. Drawing on the value chain framework
just described, consider the separate market functions of transaction
services and information provision. The idea of our web is to
provide an environment…
- where multiple transaction services providers can compete
on product and price (both quoted prices and commissions,
which may be lowered due to payment for order flow); and
- where information provision is centralized to provide fair
and transparent trading. This provides an important resolution
to the dilemma debated for years wherein centralization is
considered the ideal for providing investors with the best
price, but where competition among Market Makers (fragmentation)
is considered the best way to enhance innovation and competition.
Information is centralized in the Nasdaq quote and trade montage,
but multiple transaction services modes are encouraged. Among
other characteristics, product variability can come through
different trading modes that deal with variable types of demand
for trade characteristics such as size, immediacy, and anonymity.
Until 1997, most of the variability was supplied by variation
in the product offerings of Market Makers. In 1997, the Nasdaq
web was expanded as Electronic Communication Networks became
directly linked to the transaction service
The major changes to the Nasdaq market in 1997 were the introduction
and phase-in of the order handling rules, which began in January
and were completed by mid-October. These rules included a switch,
beginning in June, to a 1/16 minimum quote increment for stocks
with bids above $10. (Stocks with bids below $10 maintained
an unchanged quote increment of 1/32.) Roughly speaking, these
innovations lowered spreads by about 30 percent and an incremental
10 percent, respectively, likely saving investors in excess
of $500 million over the course of the
The order handling rules have two key provisions, with two
consequent economic effects: subject to a few exceptions, investor
limit orders began to be required to appear within the central
Nasdaq quote montage either reflected in Market Maker quotes
or through ECNs; and Market Maker orders placed in ECNs began
to be required to appear with the central montage. The investor
limit order aspect of the rules constituted an increase in the
directly manifested supply of liquidity— theoretically, a movement
along and down the demand curve and leading to more liquidity
provision at a lower price. The evidence, in terms of investor
savings, supports the theory. The ECN aspect of the rules had
an additional incremental effect, lowering spreads for the simple
reason that the definition of the spread was now changed. For
the first time, ECN information was included in the Nasdaq montage,
which had already been, for Instinet, a key (though unlinked)
aspect of the Nasdaq market for some time.
For reasons that are well known to students of market microstructure,
the effect of the move to 1/16 quote increments was most noticeable
for stocks that had already been phased in to the order handling
rules, that traded in substantial volume, and whose stock prices
were relatively low. This finding suggests that continual shrinking
of the quote increment will not necessarily lead to additional
Decisions made by Nasdaq in 1997 will have further consequences
for the market in 1998 and beyond. These include the decision
to move to decimal rather than fractional quote increments,
the proposal to replace the SOES and SelectNet systems with
an integrated order delivery and execution system, and the proposal
to institute a Nasdaq book for order aggregation in the Nasdaq
The web framework discussed here provides the foundation for
the idea I would like to conclude with. I believe that our greatness
will correspond to the degree to which we can express one simple
idea: openness. I see five dimensions of openness to work for.
The dimensions are mutually reinforcing.
- Open work processes within NASD, including open—in
fact gushing—information flows and openness to new ways of
organizing our work relationships.
- Openness to new content—new solutions to running
our market and its regulation. To an impressive extent, I
believe, we attempt to improve our market continually.
- Open communication—particularly as we fulfill the
potential that is available through the strengths of each
of our market constituents, through their interaction in our
market, and through their input to our improvement initiatives
informally and on various committees.
- Open market and regulatory architecture. I believe
that our finest legacy comes from our historically open market
concept. By building on the strength of our multiple Market
Maker system, driving it toward multiple-modality, we can
be all things to all investors. We can make trading seriously
cheap. We can become inevitable. This can happen in NASD RegulationSM
(NASDRSM ) too, as we learn
how to use technology in tandem with self-regulation to leverage
maximum insight into regulated-entity practices.
- Open systems. This is a technology point. It is the
technological manifestation of item four, leading toward massive
transparency and control at the very edges of the NASD network,
through PCs everywhere. Perhaps it should be combined with
item four, but I have kept it separate to highlight the criticality
of our computer systems and the people who maintain and improve
The NASD’s policies are toward openness across all five dimensions.
More and more, this is a characteristic of any successful organization
that seeks to attract and retain high-quality staff. But it
is especially true for an information processing and producing
organization such as the NASD.
From a transaction stand point, both the buy-side and the sell-side
are starting to think about different ways of doing business.
The growth of ECNs has changed the landscape in the marketplace.
The buy-side trades are getting closer to the point-of-sale.
They are influencing pricing far more than they have ever done
before. This has forced the regulators to take a look at the
structure and see if everything is working right so one can
monitor this type of activity.
There is always a fear of doing business away from the core
market. Whether it is the New York Stock Exchange or Nasdaq,
you were going to get fractionalization. Everything was going
to be done in different places and you were not going to get
everybody negotiating together.
With screen-based systems you can have indications of interest
in transactions taking place pretty much anywhere and have it
perfectly visible to most of the players. There is also the
issue of anonymity as the consolidation of order flow falls
within fewer and fewer institutions. As institutions get larger,
there is still that concern over anonymity.
There is a concern of getting the best price. In the institutional
community, you will go anywhere to (a) source the liquidity
and (b) trade at the right price. People are finally breaking
away from the thought that "it has got to be done here,"
or that "it is not the right price." You are pricing
the merchandise on your own, whether it is on an ECN, off board,
upstairs at a brokerage firm, or on a new type of system. People
are willing to trade away from what we call the central market
today. There is competition, and I think that is very positive.
We are all struggling to find one solution that satisfies everybody.
My question is, can one size fit all? If I think of it in the
clothing context, the answer is an obvious "no." If
I think of it in terms of market structure, it is also a resounding
"no." There are an unbelievable number of constituents
in the market today and in various parts of the market. When
I started in this business, things were pretty simple. We knew
what exchanges were. We knew what brokers were and we knew who
the customer was. Today, I do not think any of us know any of
those things, starting with the exchanges or Nasdaq. I am not
exactly sure what Nasdaq is today and for that matter what it
wants to be in the future. How can we get from where we are
today— which is a very complex and, for many, a confusing and
challenging environment—to a better market?
"Do we have the best market today?" Some evidence
would say we have a great market. Evidence also shows that many
of the people on the buy-side finish each day with unsatisfied
orders that at least theoretically they would have liked to
have traded that day but could not get through the process.
I, for one, would like to see a freer environment to encourage
the innovation process that has to take place. It is frightfully
expensive and it takes an inordinate amount of energy and knowledge
to ferret one’s way through the regulatory and business issues
we face every day to bring out a new system. Probably the biggest
uncertainty that we face is, will the market want what you have?
This country is so regulated, so confined in its structure,
and sometimes the regulatory environment is used as an impediment
One of my major concerns is the unintended consequences of
change. I do not know how we go about dealing with that because
it is a conundrum— how do you have an innovative environment
and not have unintended consequences? A perfect example of unintended
consequences, is where the Nasdaq folks got together to develop
the Small Order Execution System for one purpose and it became
a device that had lots of unintended consequences. I have been
a Market Maker at Nasdaq and frankly this may surprise you,
I do not think of the SOES bandits as bandits. I think of them
as people who are coming to trade at my price because I have
my price up there.
I am worried that as we go to almost an infinite number of
price increments, we lose the clustering of orders at a price
that in fact gives greater size. We may find that we have only
small orders at every price increment. It then becomes very
hard to figure out where the real prices are.
One of the things I like to do when meeting somebody from Nasdaq
is to ask, "What is Nasdaq’s mission?" "What
is it that Nasdaq wants to do?" Those are the key questions.
The term I would like to introduce into this discussion is
"bandwidth." The plummeting cost of bandwidth makes
so many things possible. These things are possible from a market
structure point of view, and they must be addressed from a regulatory
standpoint as well. This was not possible a few short years
ago. A lot of people think that Nasdaq is in the technology
business. There are private firm alternatives for members and
for investors. There are some disgruntled members that are finding
their costs are exploding. If you look to the buy-side, dealing
with one entity rather than multiple entities can provide different
sizes for different participants. That is going to provide choice
All these things can be done in a way that is transparent to
the trader. The market we think of in the future is not going
to reside in one single computer. It is going to reside on the
screens of the traders who are making the markets.
Was there in fact a need for reform? Clearly there was. Spreads
were not behaving the way they should have behaved. The SEC
set out with an admirable target of trying to tighten spreads
and to increase transparency in the marketplace. The SEC should
be pleased that the goal was largely accomplished.
There are bad apples in every market and in every industry,
and as a self-regulatory organization, it is Nasdaq’s responsibility
to weed these people out. I will contend that spreads were too
wide because of predatory practices in the marketplace. If you
look back, spreads started to widen around the time the SOES
impact deepened within the marketplace.
Spreads widened because of traders’ natural response to predatory
practices in the marketplace. Traders collectively were responding
to something they did not like to deal with, and it was somewhat
forced upon them. Where it was forced upon them gets back to
tier sizes. Tier sizes are an unnatural way to create
liquidity.(8) You have to interject back
into the marketplace natural volumes.
Customer orders are clearly dictating what the natural volumes
are out there. The Market Maker’s choice to show up at a price
with a certain amount of size, which is also natural volume,
should be encouraged. Market Makers showing up at a price with
a forced volume size does not do anybody any good.
Short of eliminating the tier sizing overnight, Nasdaq is currently
eliminating two order execution systems and consolidating them
into one integrated order delivery system. I think it will make
everybody’s lives easier and will enhance what was clearly the
goal—tighter spreads and more transparency in this marketplace.
It may become more commonplace to find Market Makers out there
aggressively trading, instead of the more benign approach of
simply trying to capture spreads. As Market Makers get more
aggressive in their trading profiles, you may interject something
in the marketplace that you clearly did not want— more hedge
fund type trading.
We must be sure that the structures we leave in place are the
structures that will continue to protect the investors out there.
In no uncertain terms, the most effective, efficient markets
with the highest degree of integrity, the markets that are the
envy of the world, are the United States capital markets. Nasdaq
is clearly an enormously important component of that.
We do need to fine tune it, however. There is nothing wrong
with tighter spreads and more transparency. We should all work
towards that end. But let’s be sure that as we get to that end,
we do not introduce things that we are not going to be happy
with three and four years from now.
Confirmation Transaction) is the trade comparison system that
supports trade reporting, clearance, and settlement.
the conference, but prior to the publication of this manuscript,
the NASD announced an agreement in principle to partner with
OptiMark and a merger with the American Stock Exchange, which
awaits an Amex membership vote as of this writing. These relationships
have the potential to extend the web (market of markets) concept
(7) At least
through 1998, you can go to www.academic.nasd.com
for detailed information on the effect of the rules and 1/16
increments on Nasdaq market quality.
size" refers to the minimum number of shares a dealer’s
quote must be good for.