consider as hurdles. and in etch to git to the point where it has to be shared. >> and you're talking about the rebate? >> correct, sir! so my question is, when you get to that point, after you say you looked at the oh factors, my question is is the size of the rebathe offered by an exchange? a knack xx in determining where you route those nonmarktble customer orders. >> yes, stl be the last factor. all things being equal, that would be a factor. >> and so the greater the rebate, that would be where you would go? >> yes. >> how many trades does ameritrade route to exchanges in a typical quarter?

>> about 37% of our flow. i'll call it 40%. >> 150,000 trades a day? >> 150,000 times a daytime also quarter. so that's going to be -- >> well, a week it would be about -- round it off. half a million a week. so may believe in a month that would be about 2 mimon. how does that sound? >> that sounds good. we did about 90 million trades last year. >> why would it be only 8 million? you asked per quarter. glo okay.

j now, we looked at your routing disclosures for the quarter tafs covered in the old paper. i'm going to have to go volt. i apologize. we're going to have to recess here for about 15 minutes. we'll give you all a chance to do something else that you need to do. what time is it now? >> 12:00. >> is it exactly 12? >> yes. >> we'll reconvene at 12:15. thank you all. >> we thank our witnesses for their patience and understanding with the is that the. let me big up where we left off, mr. quirk.

i guess we were talking about how many trades ameritrade routes to exchanges in a typical quarter. i think we rounded it off to about 8 million or something like that? >> correct. correct. we looked at your form, 606, for the quarter covered in that paper which was the fourth qualitier of 2012 tda.m.e.r.i.t.r.a.d.e. marketed that order. now, among all of the exchanges paid the highest rebathe during the fourth quarter of 2012, which is, again, the period covered by the paper and you say that the orders it was your

policy directed on the basis of best-execution. it allowed you to route all of the orders to the markt that paid you the most. i find that to be a pretty i believe credible quince dengs. >> now, you've directed all because that exchange offered your company the best execution.

the disclosure didn't send a single order to the new york stock ek change. >> no. >> and mr. quirk, how much did td ameritrade receive in rebates from exchanges last year to routing orders to venues that pay maker rebathes. do you know how much you made just from payment for order flow and rebathes? >> i can estimate it based on what where he discussed. it would be about $80 million.

>> how much did td ameritrade -- >> i don't know the answer to that question, but i can get it. >> i think it's close to zero, isn't it 1234. >> no, it's not close to zero, but i don't know what the answer is. it would not be significant. >> all right. will you get the answer for the record? >> yes. >> so, anyway, for virtually every trade, your customers, you say, were better off by your routing their orders to the exchange that paid you a rebaet rather than a venn yup that tdameritrade would have had to pay a fee. is that true? >> i would say in the sub sell kwent 24 months, you'll notice in our 6 06s that we've routed to a number of exchange ins one corner.

some of those exchanges would not be the highest rate exchanges. >> in your 606 disclosures, td ameritrade routed all disclosed, non-marketable orders to either direct edge or lava flow. the exchanges to offer the highest rebates available in the markt. is that true? >> that would be true. >> so, again, your best execution vir dhul thrill always let us you -- >> no, it would bt have always. >> i said vir chummily always. >> virtually, yeah.

>> i want to thank the witnesses for coming. what effect would banni inmake r payments have on your stock exchanges? >> thank you, senator. the effect would have us change the way we access our market. as ied earlier today, we take on the order of two credibilities for every 00 traded as revenue for conducting the informations of the exchange. so we'll simply recon figure the pricing mechanism that we have so we can continue to operate our business. it's not fundamental to the way we do business. >> we would have fewer order

types, which would reduce complexity in the market. we would likely have fewer venues, senator, as well. we wouldn't need all three of those in a world where we didn't have maker taker. >> however, i do want to point occupant it's also problemen for dark markets. therefore, we would need to kov r it with a trade app provision. >> i guess i have a question for all four of you. in your business and ours, perception reality and ri alty is perception. and i think you would agree and we wouldn't be having this hearing if it wasn't for sifg nif kant questions out there about whether you do business

fairly, unfairly, if there's favoritism, even was charged in michael lewis' book that there's real corruption. i think you would agree that there's a problem out there. would you agree with that? or you don't think there's a problem. >> well, first of all, beginning with you, mr. farley, one, do you believe there's a public accepting problem. if you think there's no pr problem out there, then just say i don't think there's a problem. bum if you think there's a problem, what do you think we aught to do? >> markets are built on confidence and perception.

>> we've proffered several suggestions. we're actually doing away with maker taker, coupling that with a rule because of some of the aftereffects reducing complexity, reducing order types, redutsing venues: when you bring that simplicity, it brides comforts. >> i think we all would agree transparency is. would elimination of that also be a positive effect or no effect.

>> mr. adam? >> in my mind, there's no question that these questions about market structure have ebb materialed into the mainstream. i think to address that, maybe a few things. e think we're seeing market forces to some extent start to do that. i believe there's three dark pool operators that have all reached prior to recredibility months not made i vailble to the public. i think you're seeing a trend.

there's a lot of insights and potentially, some additional korn tents in the markets. the medium term response is to let the fcc arctic plait to the industry, this wholistic review will be fumly corp. rehencive. some things will undoubtedly change in that process. they'll find ways to optimize and improve. there will be a decent mark. we've concluded that this is a good market structure and we're going to leave it in place.

>> there's a perception problem and a confidence issue. >> the majority of our flows have been sfwoo our equity market-based products. vanguard had 138 billion of client aspects come in the door and 76 billion in the first in this year. the best majority of those have been going to our equity products. >> well, i think it's like the old story of the guy in a small town who said to the other guy wharks are grow going to do on saturday night? e says i'm going to the poker game. he said why are you doing that? he said because you know the game issed. he said it's the only game in town.

so, mr. brennan, i don't accept your allegation that everything is fine. so if that's your view, i respects it, but i don't agree with it. >> thank you. >> i would problemly have a close review to mr. brennans. when i discussed the view, i'm going to discuss the view of our six million clients and tole you in the behavior that is they've kpibltsed in the last couple of years, it would be trading accounts increase 31%. it indicates how much exposures people are taking into market. are they participating in the rally that's happened over the last couple years. and a significant portion of them being half. that being said, i would agree with you that there is a perception problem in the segment of these clients, those would be the clients that are probably closest to this. most of mom and pop, really,

these terms don't immediate anything. they're just terms to them. the problem is in that segment, as senator johnson pointed out, that we don't spook them. we don't want them to think that they're being treated unfairly. >> thank you. i thank the witnesses. >> thank you. mr. brennan, earlier this year, you said that some high-frequency traders play a role in knitting back together a fragmented market structure. but that other high frequency traders may be unfairly taxing the testimony. now, when you mentioned some high frequency traders unattached system, were you talking about phones that engage in practice where they try to

capture rebates without actually providing liquidity. >> that would be one example, yes. >> so there are folks there which -- therefore, the maker, taker system, contributes to that problem. >> i think the markts structure is a whole. it's improved dramatically over the last 20 years. >> i'm fauking about the part that hant improved. what you just said, some high frequency traders unfairly track the system.

>> i think the maker taker in combination with differential data speeds contribute can you have arbitrage without rebates? >> no. >> so rebates contribute to the problem. >> yes. yes. mr. quirk, i think you were critical of the batallio testimony. i wonder if you'd be able to provide him with data to analyze? >> yes, we were actually asked by professor batallio after his paper was publish. i think we would be willing to

share data, of course, you know, we'd have to ensure that that wasn't going anywhere. when you were asked, what was your answer? >> to be honest with you? i'm not entirely sure. i believe that he was told that it was being considered, but i'm not certain.

>> let me conclude with saying the following. number one, we had a very good hearing, a very constructive hearing. we've heard a consistent message. and that is that there's a lack of confidence in the markts. the conflicts of interest contribute to that lack of confidence, they may lead to investors being worse off, that's what his study shows. all of these problems should be and can be addressed. one of the ways we've got to do

it is to remove the conflicts of interest. some have been very dramatic. age so, hopefully, the regulatory agencies are going to take action. ms. white, as a numg beryl of you have mentioned, said that they're going to look at structural issues, long-overdue. >> and i think there may be a role for congress.

>> we may disagree as to with what those steps are. to deal with conflicts and to deal with the other kinds of problems which exist in the current market. it's clear there can be improvements. very much appreciate your testimony. we're sorry it was interrupted by two votes of the senate. but that's the way our life works around here. i wish we could pass a law to end interupgss in hearings.

thank you, all, we'll stand adjourned. >> book tv sat down with hillary clinton to discuss her newest book, hard choices.

i rerned before to expect the unexpected. nobody expected arab spring until it was upon us. while we tried to build the world to be what we want, we've got to be aware of the fact that all these other countries. all of these billions of people making hard choices every single day. i am absolutely convince pd that we have to continue to lead the world into the kind of the future that we want. we can't sit on the sidelines. we can't retreat. we're going to have setbacks. over time, we want them to understand our story.

>> we can't abdicate our responsibility. >> hillary clinton spoke to us about her decision-making proce process, the perceptions of the united states around the globe and some of the decisions she had to make as secretary of state. >> you had broadcast tv and then cable came along. but then satellite. what if they said we're different than cable. we have a slightly different technology.

we're going to take that and not consider ourselves to be what is in law call an mvpd. so we don't have to negotiate. but satellite didn't do that. we've said from the giping, this isn't about being opposed to technology. maybe there's a business model for it. but that doesn't mean you can evade the law to run a business.

>> so each panelist will present a brief overview of their respected areas. they are in your package if you so wish to indulge. >> i thought it's probably west if i give you some examples of the research that we are conducting and the research took place with making a difference and real applications. first, i want to talk about counter measures. and then about bioterrorism risk detection portals.

>> i don't know if you remember the time when congress debated whether or not to deflect missiles. i thought this can't be a good idea. we did what we now call a defender decision approach. and we came up with a conclusion that it is very hard to make a case that mernt that's cost effective.

on the defendant of bioterrorism. that process that we designed at the time to elicit probableties on threats and successful or unsuccessful text from primarily intelligence agencies is now standard use in dh zrvegs and is combined in other risk assessment procedures, for example, chemical and radiological and knew lar. >> currently, we're working on evaluating advanced methods for radiation detection. we did a previous study on the advanced radiation portal which showed it was not a good idea to replace all portals with the new ones, but to perhaps use them in secondary inspection. and that is still under investigation.

i think the philosophy is we're very much relevant behind it. we're picking problems that are relevant to the problem. in that process, we always invent new things in terms of methodology and tools. it's a give and take chlts. >> so i'm going to talk about computational game theory. the question we are faced with in our research and we're very thankful for the creations that were brought and the partnerships we've formed with many of the security agencieage

which allows us to conduct some of this research. so game theory is a wobdserful tool. what's unique about thework is it blends computational and behavioral game theory. what's also interesting is the opportunity we've had to create to deploy this using game theory in the field allowing us to get data back. from the field. so these are things that are unique and this is also possible because of all of the different agencies that have helped to fund.

these are very important partnerships. this has also been used for the staten island ferry. you heard today about the isis system. this is for assigning air marshals to flights. many students are sitting right there at the back of the room. this has been a fantastic partnership. this was also the program that was initially deployed. more recently, what we've been looking at, our applications where there is a loot mr

interaction with adversaries. we're also looking at patrols in the gulf of mexico for wildlife security against poachers. there's a lot of interesting work that we can do here. and the agreement uses inspired perspective. the idea that basic and applied research are not in opposition to each other, which really go hand-in-hand which has been crucial to allow us to kukts this research. >> bill, please? >> i'm going to focus on our

team'swork in the area of how the public responds to threats and hazards. i want to thank you the fellowship that i did last supper. this is important. came away with an idea of how our science can be a better contributor to the mission of not only great, but dhs in general. so i'm very happy about that. if you think about the remarks that we heard in various talks

and the previous panel, every one of the speakers noticed a pivotal role of how the public responds to risk. the public is a major player. we've been able to document the first time the economic impact of how the public responds. now, we knew this anecdotally. and if we think about how people reacted after 9/11, there was a propensity not to get on airplanes and so forth. but we grew upon some of the best social science folks 234 the world, quite frankly, brought them into the team to apply best social science to understand this problems so that we could understand how the

public responds. i's very important if you think about it. and i echo malin's comments and detlof's comments about how we allocate resources. the oh part of that equation is the risk communication dimension. as we have seen in rekrents years, tremendous leverage to mitigate the public. the public doesn't always overreact. in some cases, in response to natural disasters, they might under-react and we design risk communication to address that. but mostly, the economic impacts come from overreaction. so we have looked at a wide array of different hazards.

a number of people designed a dirty bomb attack about three miles from the financial district, simulated attack. and we engaged our nationwide panel to see how the public miegts react to this kind of highly disruptive event. i hope i'm not stealing your thupder here. about 1500 of how the public was likely to respond to an event like this. we have developed capableties anticipate iing how the public reacts. some would say they're very compelling chltsd some would say they're even scary.

to better understand how they might respond in the future. we were on the field the day the stock market collapsed. what does that have to do with terrorism? we were able to understand how quickly people recover from crisis. that was very important. we were in the field for days. because we were in the field, we were able to understand how people responded to the haiti earthquake so we could compare natural disasters with the terrorist attack.

we have gained a lot of understanding. we conducted a risk perception study for the tsa starting about a year and a half ago, where we worked very closely with the tsa to target some attacks. we were able to document for the very first time the economic value of pre-crisis communication, communicate with the public in advance to bolster confidence. and the -- as adam mentioned, that we puts the value at about $400 million of prevented loss revenue to the airline industry for a risk communication project that probably would take under million dollars to execute.

there's huge leverage there. we have dwoched the capableties now. we're fast-moving to actually take computer science to help us in code what has really been a lots of behavioral science pioneered and we've extended. we're now able to code that and create computer simulations. one of them, the most flejling version right now is back. you can see that. and then most recently, we're very, very interested in social media. both as a conduit to really understand in realtime how the public is likely to respond to variation threats. but, also, as a way to test our -- and communicate and diffuse our mes sanctions. >> great, thank you, bill. >> so there you have it.

>> adam? >> well, i would like to summarize how we've molds earnized what used to be called hazard loss evaluation. we've done this by including several important features. the first of these is resilience. we've developed the effectively to engage the key feature of the new paradigm. this is best kpemplly filled by the fear factor.

fear does have a cost. we found this swamping many of types of koost. we've included remediation in this framework. there's a big debate about the money become spent on an an tlax attack represents a cost or a benefit. and rather than getting em brawled in that debate, we simply look at the effect of what effect it has on gdp. we also have included mitigation. now we can do benefit cost analyses. and an important, new innovation has been to examine spillover effects of counter measures.

they may have an effect on the business envooirnt. it turns out that's not always negative. more police officers, more cameras make many people feel more comfortable and can actually have a positive spillover effect chlgt these are very significant. >> all of these are with creative affiliates of other oup centers and with dhs operating agencies. let me just take a minute and talk about how these methodologies have been applied and implemented in dhs agencies. the first one was the risk assessment where columbus developed this framework for doing consequence analogy sis in an automated system.

so it could be applied to hundreds of scenarios. currently, i'm working on expanding the frameworks being used at the domestic nuclear detection office in terms of getting better data on the duration of the time path for these con kwenss and for the national biosurveillance inform vgs center in terms of broadening their framework to include this comprehensive set of impacts. these tools that have been used on late times and urban kmers and security study which examined the trade-off between those two. yes, the delay time and the

inconvenience may be small per person, but when you talk about the city of 10 million people, it quickly magnifies. finally, we've applied these to a lot of unconventional types of cases, such as an h 1 n 1 epidemic attack, chlorine attack and many natural hazards such as erts quakes and tsunamis. >> thank you. >> first, let me just say how delighted the award and risk center has been to have interacted here. as you already know, the work that we are doing at the risk center has focused on the opportunities that insurance prevents to try to reduce the risks that one faces. terrorism certainly being a

critical one. and we focused particularly in recent years on the flooz had ard, which has become a major problem for us. we intended to use very much the framework that you outlined at the beginning, isaac, in terms of the perception in which you've heard from bill along with risk assessment and risk analysis and risk management. and try to pull them together and try to figure out a way that they can be implemented. a framework that we have utilized to get these concepts across comes from a book that many of you have read, thinking fast and slow by daniel connaman. we see these two aspects, we see intuitive thinking that plays a key role here. we're all short-term oriented, all of us.

we are also trying to figure out ways to be more deliberative. let me give you a couple of examples as it applies to the area of flood insurance which we have been looking at the last few years. we've been doing experiments on this with a plot of other people here in this room. pete and isaac alluded to this. here are a few cases with respect to insurance. people do not buy insurance until after a disaster occurs, unless they're forced to do it. voluntarily, there's a tendency not to buy it. they'll purchase a policy for a few years.

but if it turns out that they don't collect or they haven't suffered a loss, they then say to themselves, hmm, this is a very bad investment. look at all the money i could have used in my premiums. and they'll cancel it. how can you convince people that the best return on an insurance policy is no return at all. you should celebrate the fact that you haven't had a loss. that has been one of the challenges that one faces. so in the process of trying to deal with that, what we have really tried to suggest is ways to develop long term strategies with short term incentives. insurance, if it is priced correctly, based on risk and we have heard the notion that we

should have risk-based. if you have risk-based pricing and you take measures to protect yourself, you should get a premium reduction. if you can tie the loan with a long term, you may actually be better off that. 's a direction that wefr going in. let me raise one challenge as i trying to figure out ways that we could somehow find policies that would be effective. my colleague is here and is mod rating the next session and has had a whole group of creativity. this was a joint effort be many, many associations and societies.

the flood insurance act was passed in july, 2012. and you heard a little bit about that briefly mentioned this morning. by adam. that act actually was revolutionary because it said insurance premiums should reflect risk but we also have to recognize there are challenges with affordleability if the premiums are too high. this particular bill was passed three months before hurricane sandy. our guess is it would never have been passed if hurricane sandy had occurred in july. but hurricane sandy did occur and this past march the bill was not necessarily repealed but revamped, so to speak and we're now facing a challenge with respect to how you're going to have premiums reflecting risk because affordability dominated the issue appropriately so. so what we're faced with now, how do we deal with the short term challenges and how do we deal with issues like

affordability so we can use tools that could be effective for therebyive thinking recognizing intuitive think cigarette often driving the kind of decisions that get made and we're facing this as of march how do we bring afford jblabili into the picture. >> the hallmark of the work that all five of you have done is that you've had -- you've done research impact which is academic, has to be in order to be done at a university but also had impact in the field. so in the audience we have folks from other academic institutions that to know how it is we indict and there may be other potential partners that are saying how do we work in such a way to enable these kinds of successes to also occur with our agencies. so what i would like to do is cover the general topic of how do you bridge the gap between academic research and operational tool development and

you had this idea right from the beginning, i remember when you were interviewing me and when we were first talking the first couple of years you recognized that this kind of research couldn't have a ten year horizon, it needed to have some near term application. maybe you can tell us a little bit about your thinking of how do you bridge this gap, when you're brought in as an expert, how do you balance between listening to the client, find out what their needs are and advising them and recognizing the challenge is that you may not be familiar precisely with what they are doing. what do we know. how did you address that in those early years. >> you characterized it well. i do a lot of listening. i certainly don't do any selling. i don't want to sell tools. i don't want to sell analysis. i don't want to sell any particular technology. i want to help. i want to make a difference. i want to be useful to whoever is the decision maker or client. that's what i said earlier.

i would like to start with a problem, not with a methodology. and at that methodology to the problem was search for the appropriate methodology and finally keep an eye on the decision that has to be made. whether that's a resource allocation decision or funding addition or technology investment decision because often people obsess about getting the correct data when, in fact, the decision can be solved with relatively simple tools and relatively simple sets of data even if they are not precise. so i think that's -- one thing i would like to say. in every single application, which may be even in the beginning appear relatively simple there were interesting research questions that came out that led to new developments and insight. >> great. thank you. now, i have seen this in many meetings i have seen with you. what was your biggest challenge

in conveying the results of academic analysis to an insight for people in the field who have their own view of what's the right thing to do. identify been to many meetings we start out with other folks sitting there what do you know about what we do. how do you convey your results? >> so, one thing we have to recognize is that these are different disciplines and computer scientists are communicating with people who are experts in other professions and somehow we have to have this dialogue. now, even within, you know, computer science and psychologiy. but one of our collaborators is a professor of psychology. when we had our first meeting and to build these applications, we needed his insights.

we would have these meetings and the meetings would go on and look at the data, and, you know, after a while we were just confused. you know, how come he's not saying anything about how to speed up our algorithms. it dawned on us he's a psychologist and this is not his primary concern to speed up our algorithms. that's not his goal. that's the kind of thing we have to learn that when we start working with our partners, that we may be very excited about our latest algorithms and so on and so forth but what we need to be careful -- and now turn it around in many instances it is their communication to us that has led to some very interesting new pieces of challenge. i'll give one example. so in 2010 or so when we had

already built a system we figured out a way of speeding up our algorithm as long, a the main complication is that air marshals and stewards make everything complicated. the air marshals would get on the flight and the alvaren algo would run very fast. the answer sue are too obsessed with fast algorithm. that's not the primary concern to us. and so, these are things that we have to learn and over time, you know, this partnership has really worked for us. >> great. we have many law enforcement personnel here who are listening to our advice here, and we're going to be on c-span. teen political aspects of this, i noticed congress folks have already left but howard i want to ask you what are the political challenges in developing these long term strategies? >> well, i think there are many

and we can probably all resonate to this. we have a little acronym, since you mentioned political challenge that we often use as a way of responding to that question and some of you may know this from the audience. nimtoo. we prefer not to do it. one of the political challenges and the things that we really are faced with is how do you present information in such a way that people will pay attention. i think there are just two points i can mention that people can resonate too. one of them stretching a time horizon, somehow letting people know what are the consequences of something over a long period of time. so if you tell someone there's a 1 in 1,000 or 1 in 100 chance of a terrorist attack or flood attack or something happening there's a general tendency to say it won't happen to me, i don't have to worry about it,

it's 1 in 100 chance next year. if you can stretch that to saying if you believe the probability will remain the same over time, 1 in 100 it's 1 in 5 chance of one of those events happening in 25 years and it's interesting people pay attention when you stretch it. the other example is focus on the worst case scene jobs focus on something that somehow says look if you aren't protected, if you don't take these measures today and you don't have some basis for doing it here's what could happen to you in terms of the consequences. so rather than focusing on the low probability, high consequences. this is a challenge, obviously, when you have the short term kind of concerns but at least it's a direction maybe that there will be more attention paid to these things. >> great. thank you. you have broadcasting and then cable came along and then satellite. what if satellite had said, you know, we're different than cable, we have a slightly different technology, we'll take

that and not consider ourselves to be what in law is called an mpvd. we don't have to negotiate. satellite didn't do that. why should aereo be able to come up with a different technology and say we don't have to negotiate for copyrighted material. we said from the beginning this isn't about being opposed to technology. there's still a technology there in aereo and maybe there's a business model for it but that doesn't mean you can evade the law to run a business. >> more about the supreme court decision against aereo with the head of the national association of broadcasters, gordon smith tonight at 8:00 eastern on the communicators on c-span 2. now you can keep in touch with current events from the nation's capital using any phone any time with c-span radio on audio now. simply call 202-626-8888 to hear congressional coverage, public affairs forums and today's

washington journal program and every day listen to a recap of the day's events on washington today. also hear audio of the five networks sunday public affairs program beginning sundays at noon eastern. c-span radio on audio now call 202-626-8888. long distance or phone charges may apply. coming up next on c-span 3 american history in primetime with tonight's focus on hollywood directors and the world war ii films they created for the u.s. military. first frank capra's film "your job in germany" that was used as a training tool for troops occupying germany after its defeat. then a look at george stevens documentary "the true glory" about the d-day invasion of normandy. followed by a film on the battle of midway from filmmaker john ford. later john houston's film on the 19