Transcripts For CSPAN Hearing On Impact Of Artificial Intell

Part of the his is Financial Services industry. The chair will now recognize an elf for five minutes for opening statement. First off, thank you for joining us today for what should be a thisinteresting hearing of task force. Today were looking at exploring how a. I. Is being deployed in Capital Markets from automated Portfolio Allocation to Investment Management decisions. Were also going to consider how the use of this technology is changing the nature of work in rendering ervices, some jobs obsolete and changing the skill sets needed to excel in others. It would not be much of an exaggeration today to say that quite literally is run by computers. Of those are the days screaming on the new york stock pour ge and they would over the ticker statement and Financial Statements to gleam value. Into a companys i actually hear about those days from the limo driver who takes e back who used to be a floor trader on the merc. Today, floor trades are orders are d executed in microseconds. E. T. F. s rely on lgogh rhythmic models to make sure its properly waved to whatever benchmark is tracking. Quan tative funds scour all sorts of market data to find the price that have the most momentum or highest dividends or look for correlations that and in in the market external data feeds to provide the most value for investors. Notable nk its very lot of the shakeup were seeing in those markets is really a reflection of sort of take all nature of digital economies. Hat any Digital Business, purely Digital Business is a more monopoly. As it becomes the distinguished advertised, you will see rewards go to a smaller number of dominant players. Emphasize its not evil. Its simply a natural reflection digital ture of the marketplace. Other Asset Managers may use complex s to perform research and analysis in real time on big data sets. Scouring social media sites, web traffic, online transactions, and just about of. Hing else you can think this is, i guess, good in terms of having the market reflect all known data, but there are abuse of corners. Imagine what it would be ert if you had a 10second early look at trumps feed how much money you could make trading off that, for example. Computer types of managed tuck fundz make up about approximately 31 market. Public equities hedge funds, other mutual funds, manage just 20 of the market. Socalled the computerization of our stock markets has a number of benefits. Trades has executing gone way down. Sometimes to zero dollars. In theres more liquidity the market. Passive funds charge less than of assets while active managers often charge 20 times that much. It creates additional questions, however. As in the 2010 flash crash and the more recent mini flash shown algogh hythmic trading can provide unpredictable consequences and between e information different types of investors. As firms with more and faster to data sets can obtain a competitive advantage. Another broader question is how impactinglopments are the nature of jobs in the Financial Services industry. Wells fargo Research Report estimated that technological efficiencies would 200,000 job cuts over the next decade in the u. S. Banking industry. Hile these cuts will certainly affect back office, call center, and Customer Service positions, widespread. L be many Front Office Workers such s bankers, traders, financial analyses could also see their head count drop by almost a hird, according to a report released earlier this year. The report also found that 40 of existing jobs at financial automated with current technology. To first order, if you spend day staring at a big screen and in particular if you your e a large paycheck, job will be at risk. Understanding the skills that will be needed to excel in the Financial Services industry of and how we can encourage these skills is one of the issues that we must tackle tackle early. In a world where many functions can be done by automated a. I. Does that t role leave for humans . So i very much look forward to hearing from our witnesses on issues. With that id like to recognize the task force Ranking Member, y friend from georgia, mr. Loudermilk, for five minutes. Mr. Loudermilk thank you, mr. Chairman. Want to thank each of our witnesses here today. Thank you for taking time to be here to discuss this issue while is fixated america on other things going on here. Something that may not resonate on the major networks, veryt is something that is important, has an impact on our lives, positively but also negatively. Its important we be looking into this. Task you know, today the force will examine the intersection between technology and the Capital Markets. Years, there have been many technological developments, including the adoption of and icial intelligence automation that have redefined nd reshaped trading and investing. The first is on the new York City Stock Exchange were made in manual 1700s using a paper intensive process. For many years, buyers and communicated about numbers over the phone. Today, trading and investing are one on digital platforms and investors can trade securities virtually anywhere in the world technology. It has benefited the markets in many ways. For investorstive by leading to lower overhead and transaction costs which to record investment returns over the last decade. Several major Asset Management offer 0 commissions which means investors can buy and sell stocks essentially for capture more of the growth of their investments. This would not be possible trading. Lectronic Digital Trading platforms also provide investors with access to lowcost reference and advice roboadvisors. Electronic trading makes markets more efficient by allowing prices, arches for better processing of large sets of data and price information. Of technology on can also lower firms barriers more ry, foster competition, improve risk management, and increase Market Access for investors. Addition to these core benefits, there are many other cases of Companies Using a. I. To improve efficiencies in the Capital Markets in unique ways. Or example, some Clearing Companies are using a. I. To optimize the settlement of cybersecurityance and fraud detection. Some selfregulatory using ations are also a. I. In reg tech and market surveillance. While there are many benefits to electronic trading, it can challenges. One challenge which is at the forefront of our discussion is the disruption of job market. Rise of automated trading has displaced many floor traders, Job Opportunities in fields like code writing, cloud telecommunications, fiberoptics and Data Analysis are growing. Here is some concern that High Frequency trading can contribute to volatility, but new evidence uggests that high frequent trading does not increase volatility and can actually improve liquidity. Is also some concern that firms dont have the latest firms that dont have the latest technology could be competed out of the markets. In mind rtant to keep that not all types of electronic trading are the same and i look forward to learning more from witnesses about the differences between automated rading, algogh rhythmic fraying algorithmic trading. And also looking at the issues in this space. Of source protection code. Because algorithms are a core intellectual property, they must be protected. Passed a out of we bill out of this committee and the house on a bipartisan basis the congress to ensure that securities and Exchange Commission issues a subpoena before obtaining these rather than getting through routine exams. We can work, i hope together on a bill this ng co. Back. K you and i yield mr. Foster thank you. Today were welcoming the provost and professor of media, culture and n. Y. U. Cation at deprado, cornell university, chief investment fficer of true positive technologies. Ms. Rebecca fender, c. F. A. , of or director, future finance, chartered financial analyst institute. Mrs. Wagner, chief executive officer modern markets initiative. Racial, head of nasdaq surveillance, nasdaq stock market. Our oral testimony will be limited to five minutes and without objection, your full written statement will be made part of the record. Doctor, you are now recognized for five minutes to give an oral presentation of your testimony. Foster and Ranking Member loudermilk, thank you for inviting me to testify. While my written remarks cover four key areas, my oral remarks two the implications of automation on the workforce mitigating algorithmic bias. We are concerned in the Financial Services sector. First, the Financial Services ripe for automation and algorithm. The in tech sector is on rise. Third large number of workers will be displaced in the Financial Services sector even automation and a. I. Development is protected to create new types of jobs. Cilwain if all of this is true, then the cause for concern for clear. Fact that h the africanamericans and latin workers, in particular, are underrepresented in the Financial Services sector workforce. Africanamericans, hispanics, and asians make up only 22 of he Financial Service industry workforce. Africanamerican representation in the Financial Services sector senior entry level and level jobs declined from 2007 to 2015. Less than 3. 5 of all Financial Planners in the u. S. Are black latin. Africanamericans make up just 2. 9 of hispanics just the securities subsector. Asians make just make up just 2. 8 of the central banking and insurance subsectors. My point is simple. Groups that are extremely underrepresented in the Services Industry will be most at risk in terms of automation and the escalation of fin tech development. This is especially true given vast underrepresentation of africanamericans and latins in the workforce. If we are to mitigate the likelihood that automation will disproportionately and affect those already underrepresented in the we ncial Services Industry, must plan ahead long into the future rather than allowing the market to run its course towards predictable outcomes. Deterring subject of algorithmic mic bias, we need to deploy these more s and provide oversight from industry, overnment and nongovernmental bodies to see what outcome they produce. Make algorithms more mitigate such potential bias rather than their effects ct once their damage is done. But i want to emphasize that specially when it comes to mitigating the potential outcome rhythms have on communities of algorithms of color, munities technologists is not a complete solution. Drawing on a by former rights leader who had a understanding of computerized automation as they existed in his time. The unskilled and semi skilled worker is the victim but cybernation invades stronghold of the american middle class as once proud workers began sinking into the alienated underclass. Meet the new r poor, we know that automation is the curse but not the only curse. Problem is not automation but social injustice itself. Financial example the findings from a recent National Research economic study entitled discrimination in area. N tech their researchers taught to an \gogh rhythm system could hurt. The findings were mix. Yes, the system discriminated less than the traditional process but also meant that the process discriminated against a number of africanamericans and latin loan applicants. Even though the system did not balance discriminate in terms of loan approval, it did latin inate black and users in terms of price. One of the key conclusions of fin tudy states that both tech and facetoface lenders may discriminate and mortgage through pricing strategies. We are just scratching the surface and the role of pricing the egy discrimination in use. Ithmic area of data ending may discriminate less. But it does not eliminate it. Ven with the aid of a fair, accurate, and transparent ystem, Racial Discrimination persists. Thank you, again, allowing me the opportunity to contribute to these proceedings. You. Oster thank mr. De prado. De prado thank you for to be here today. Oreign tasks until recently, only expert humans could accomplish. One is management of investments. For two reasons. First, one of the most successful things in history to be algorithmic. He key is their decisions are objective, can be improved over time. Automation enables cost reductions. Including order forecasting, other things. Number l a. I. Creates a of challenges for those involved in the financial and insurance industry. Many of whom will lose their jobs, not because they will be replaced by machines, but because they have not been alongside work algorithms. The returning of these workers important and difficult task. Not everything is bad news. Skills become more the genders, etween ethnicities and other classifications should narrow. A great equalizer. Returning our existing workforce importance. Al however, its not enough. We must make sure that the that american universities have contributed and developed remains in our country. Founders of the next google, very , or apple are this morning attending a math or ngineering class at one of our universities. These future entrepreneurs are in our country on a visa. Unless we help them, they ill return to the countries of origin. And they will compete against us. On a different note, id like to attention to two practical examples of the learning n of machine algorithms to regulatory oversight. Crowdsourcing of investigations. One of the most challenging to s faced by regulators is identify market manipulators data. This is a challenging task. Like searching for a needle in a haystack. The practical approach is for help ofrs to enroll the those. Regulators could minimize transaction data and offer to the Worldwide Community scientists who will be rewarded with a portion of the finds. Time that Financial Markets Something Like the couldcrash, this approach lead to faster identification of potential market manipulators. Second embodiment of reg tech, they are filled with false a Financial Firms even online tools and large hedge funds fall constantly for this trap, leading to investor losses. To require is Financial Firms to record all those involved in the development of the product. This information, say its could overfit and this probability reported. Finally, i would like to conclude my remarks with the bias. Sion of yes, Machine Learning algorithms can incorporate human biases. News is we have a in er chance in protecting algorithms. The reason is that we can algorithms to a batch of andomized controlled experiments and have them perform as intended. Can assist human ecisionmakers that humans can override. Hus and as algorithmic investi investing goes on, many can reap the benefits of this technology. Thank you for allowing me to contribute to this hearing and i look forward to your questions. Mr. Foster thank you. Ms. Bender, youre now recognized for five minutes to presentation of your testimony. Foster, r chairman loudermilk, and members of the task force, thank you for inviting me to testify here today. My name is rebecca fender and i senior director of the future of finance at c. F. A. Is our thoughtch leadership platform. C. F. A. Institute is the largest association of investment professionals in the world with 170,000 c. F. H. Countries. Ders in 76. F. H. Is best known for the Charter Financial analyses which graduate level exam. They must have four years of experience. C. F. A. Institute is a nonpartisan institution and voice on e a leading the globalization of market protection. C. F. A. Published a paper on the of the future, examining the changing roles in five to 10 years. Among the c. F. A. Institute members and candidates we 43 think the role they perform today will be substantially different in five to 10 years time. Was greater than 50 among financial advisors, traders, analyses. Another 5 do not think their role will exist by then. One of the catalyst is technology. Its a pyramid. T the foundation we have basic applications. Everythi everyone will need to learn to and some differently eople will face tech subis i constitution. Substitution. Technology will enhance work. Are e top there hyperpartisan roles. Believes the te key is ongoing learning. Curriculum now has Machine Learning. And among those surveyed in the interestport, 58 have in Data Analysis coding r. Guages like python and similarly, Data Visualization and interpretation are areas expressed han half from in. In terms of the role of Artificial Intelligence in the nvestment industry, the organizing principle we see is Artificial Intelligence plus intelligence. Or a. I. Plus h. I. In these middle and top levels hierarchy, y Investment Management justice and Technology Teams Work Together. Augment niques can human intelligence to free and tment professionals enable smarter decisionmaking. Markets ensure how work. It unlocks unstructured data and identify pat