1.
몇 일전 치타와 슈퍼메뚜기라는 글에서 CFTC 의장인 Mr.Clinton의 비유를 소개하였습니다. 연설중 HFT를 언급한 원문입니다. 미국감독당국이 HFT를 바라보는 시각입니다.
Cheetahs?Cutting the Queue
Technology is the first area where changes are occurring at breakneck speed. You already know that since you’re attending this HFT conference. Technology can be a great equalizer, bridging people across oceans, between rural and urban and rich and poor. However, there will be a steep price to pay if regulators and exchanges around the globe don’t effectively manage the change taking place as a result of new computerized trading. In financial markets, folks screaming at each other in trading pits have quickly become mostly a thing of the past. Instead, computers are screaming at each other all day and all night?most times regardless of time zones around the world.
High frequency trading (HFT) does add liquidity. It adds access. Where do you think the third largest trader by volume on the Chicago Mercantile Exchange (CME) is based? In Prague. Now, that’s access that wasn’t there ten years ago. For us regulators, technology also provides an electronic data trail. At the end of the trading day, exchange employees used to scoop up the little tickets on the trading floor with snow shovels and that’s the data we used many times in our enforcement efforts.
It is amazing how quickly these markets morphed. In the U.S., well over 90 percent of the trading is done electronically. HFTs alone account for roughly 50 percent of the trades in Europe and roughly a third of the trades in the U.S.
I have a moniker for high frequency traders: cheetahs. In the animal kingdom, cheetahs can run seventy miles-per-hour. They’re the fastest animal on land. Zero to sixty in three seconds?now that’s fast, fast, fast. So are markets today.
The cheetahs are out there nearly 24-7-365 trying to scoop up micro dollars in milliseconds. We need to do a better job of keeping up with the cheetahs. Legislatures and regulators alike need to be more nimble and quick, anticipating the nature and pace of market innovation and change.
Here’s one of the potential problems I’ve been hearing about. These markets started, in large part, so that commercial entities could hedge their legitimate business risk. Today, too many of those folks?farmers, airlines and others are telling me they can’t get into markets at their price point because they get edged out by HFTs, by the cheetahs. Like when a person heads to the line at a buffet, the cheetahs cut the queue and jump in front of the commercials.
I’m also concerned about allocation algorithms, that some of the exchanges may use. These may be adding to the problem by not necessarily accepting the first or best bid or offer but weighing the size of the trade, too. From an exchange business purpose, I get it. More volume equals more money for the exchanges. However, cheetahs may be gaming the system by bidding or offering more contracts than they believe will be filled simply to cut the queue to get ahead of other traders. They may receive an advantage and then have their order partially filled. There is every reason to assume that a cheetah’s algo program could determine the size of the order that would allow them to cut the queue. Regulators have, to date, simply accepted that all is well with how technology is working. That needs to stop. We need to be more inquisitive. We need to think about these kinds of things before they reach trouble points manifested in market anomalies.
In the interest of transparency, I hope that we hold a public meeting or hearing on these allocation algos and other specific issues related to cheetah trading. But, we aren’t the only ones who should be looking at these matters in public. These are global multi-market matters and others should consider what they might do to not only better understand, but inform the public about the impact of these technologies.
Now, I’m not here to criticize the way the cheetahs do business. And by the way, some of them that I have spent time with are super impressive. As a regulator, though, I do want to make sure these markets are safe for users?including cheetahs.
We don’t have to look very far to see where problems have already existed, such as the Flash Crash. We recently received recommendations from an advisory committee which provided us with some thoughtful suggestions about it.
Already, circuit breakers have been put in place in some securities markets, but they need to be expanded and they need to be harmonized with other U.S. markets so we can stop the kind of arbitrage that created a cascading affect across all markets. Should these types of circuit breakers be harmonized in other nations? Maybe. Think about it. There are stocks and futures which are arbitraged internationally. Had the Flash Crash taken place in the morning instead of the afternoon, the impact could have been worse because European markets would have been open. Since it took place in the mid-afternoon, it was primarily limited to U.S. markets. We have greater coordination and controls in place both in futures and equity markets, but we still need greater harmonization between markets, both in the U.S. and abroad.
Another recommendation is for trading programs to have some kind of “kill switch” that could be activated when a program is feral. Most of the time, it’s innocent. Even so, there’s the possibility that these cheetahs can roil markets and that’s what regulators need to get our heads around. If trading becomes uncontrollable and roils markets and costs people money, there needs to be accountability. That just makes markets safer for everybody.
I also believe these trading programs need to be tested, probably by the exchanges, before they go live. At the Intercontinental Exchange?London, they do conformity testing. In India, regulators test cheetahs to ensure they don’t have the potential to violate fraud, abuse or manipulation rules or regulations.
Still, it would be na?ve to think there won’t be glitches. That’s why I think additional safeguards may need to be put in place after there’s been a problem. When a plane crashes, for example, the airlines reprogram their simulators to create the exact circumstances that led to the crash so that pilots can train to avoid a future problem.
I was reading this article about airline safety that included this?quote: “as an industry, we need to improve upset recovery training.” The?same is true in markets. We need to improve our “upset recovery?training” with regard to algos and cheetahs.
CFTC commissioner: High-frequency traders, algos are concern중에서
한국과 너무나 다른 미국입니다. 비교할 수 없습니다만 그래도 대한민국 금감원이나 금융위의 자본시장 담당책임자가 HFT등을 놓고 연설하는 모습을 기대해봅니다. 신문기사에 “관계자에 따르면”라는 식으로 짧은 문장 말고 길고 긴 철학을 듣고 싶습니다.
2.
만약 물소 몇마리외 수 십마리의 치타들이 드넓은 아프리카 평원위에? 있으면 치타들은 생존할 수 있을까요? 치타들은 생존경쟁때문에 새로운 탈출구를 찾든가 아니면 굶어죽던가 갈림길에 놓입니다. 수퍼메뚜기만 득실거리는 시장에서도 마찬가지입니다. 만약 ELW수사가 일어나지 않고 증권사 VIP서비스가 계속해서 늘어났으면 어떻게 되었을까요? 결국 ELW스캘퍼들의 수익율이 떨어졌을 것으로 예상합니다. 아무리 Latency를 줄인다고 하더라도? 임계치가 있기때문입니다. Zero To Latency라고 하지만 어느 수준이상은 힘듭니다. 몇 마이크로초를 당기려고 수 십억을 투자할 수도 없습니다.? 이런 상상을 연구하여 보고서로 낸 곳이 있습니다. 유명한 Tabb Group입니다.? 주식시장의 50%가 HFT인 미국에서 HFT가 임계치에 도달했다는 보고서가 나왔습니다.
QUANTITATIVE RESEARCH: THE WORLD AFTER HIGH-SPEED SATURATION
임계시점에 도달했다고 주장하는 근거로 속도를 줄이기 위해 투입하여야 하는 비용이 늘어나고 감독당국의 규제가 높아진다는 점을 들고 있습니다.
Like a good mystery novel, evidence is mounting that an era overly?focused on high frequency trading is coming to an end.? It might be?enough to draw this line in the sand based on prop shop closings,?increasing incremental costs of speed, and regulatory efforts to?manufacture friction ? through rulemaking and fees – in an otherwise
frictionless playing field.? But, on top of this, there is both direct?and indirect evidence that much of the high-turnover world is actively?seeking new forms of alpha around which to deploy new strategies.
사실 임계시점인지 아닌지를 떠나 미국과 유럽의 HFT들은 임 2010년중반부터 새로운 시장을 개척하기 위하여 남미나 아시아시장으로 발걸음을 옮기고 있습니다. 나아가 주식을 넘어서 새로운 투자상품을 찾기 위한 노력도 이루어지고 있습니다. 그중 가장 큰 각광을 받고 있는 상품은 FX입니다.
그럼에도 불구하고 HFT를 통해 막대한 이익을 챙긴 트레이딩그룹들이 찾는 새로운 모델은 무엇일까요? 보고서는 다음과 같이 정리하고 있습니다.
With these drivers as a backdrop, TABB Group believes that we are?embarking upon a new era of quantitative ? research, strategy development?and other targets of trade workflow automation that will usher in a?period less focused on speed and more focused on hunting for patterns in?a much larger and diverse sea of financial instruments, market-related?data and use cases.? Though speed will remain important, particularly?for electronic trade execution and as regions such as Asia-Pacific catch?up with the West in terms of market infrastructure, it will no longer?represent the dramatic competitive advantage of the recent past.
The trade strategies of this new era will undoubtedly be much more multi-dimensional.? By default, quants will need to don their best Captain Kirk impressions and “go where no [person] has gone before”.? And like any journey into uncharted territory, old boundaries will need to be tested.? Among these will be a greater emphasis on cross-asset, cross-regional, and multi-temporal strategies; a new category of trade strategies that we call “multi-dimensional arbitrage”.
Multi Dimensional Arbitrage Based On on cross-asset, cross-regional, and multi-temporal strategies으로 정리할 수 있습니다. 이런 전략은 유명한? 스타트렉의 선장인 커크의 표현처럼 미지의 세계를 개척하여야 합니다.
Go where no [person] has gone before!
FT Alphaville은 위의자료를 인용하면서 다음과 같은 그림을 그리고 있습니다.
The strategies, we imagine, will likely focus on detecting arbitrage?opportunities between US-listed stocks and ETFs in out-of-hours trading?versus the underlying stocks and futures listed on international?exchanges. A fact facilitated by the recent proliferation of?cross-border exchange tie-ups, cross-listings of major index products in fresh timezones and ongoing regulatory under-sight of a cross-border trading world.
It also means applying high frequency strategies to 24 hour markets like commodities. And arguably makes the scenario of a so-called splash-crash ? a multi-asset flash crash ? an ever more likely possibility. At least according to us. No wonder there’s been so much activity in the field of manufacturing arbitrage in the emerging markets sphere of late.
The future is all about cross-asset arbitrage중에서
그러면 아시아시장은 어떻게 될까요? 아래와 같이 정리하고 있습니다.
That includes applying high-speed trading strategies to emerging and?frontier markets ? with Asian markets currently considered one of the?biggest of opportunities and bounties to be had.
3.
위의 보고서에 따르면 HFT도 트레이딩 흐름중에 나타난 한가지 특징이라고 정리할 수 있습니다. 그렇지만 HFT를 가능하게 했던 Low Latency Technology와 Machine Driven Trading이라는 거대한 흐름을 벗어난다고 생각하지 않습니다. 패러다임은 변화하였고 변화를 딛고 수익율을 높이기 위한 새로운 전략찾기로 이해함이 타당하지 않을까 합니다. 그렇지만 새로운 전략이 당장에 나타나지 않을 뿐 더라 아직도 HFT가 유효한 신흥자본시장때문에? 몇 년이라는 시간이 필요하지 않을까 생각합니다.
한국 자본시장이 어떻게 변할지 지금 당장 예상하기 힘듭니다. 파생상품시장은 50%이상이 HFT거래라고 할 수 있지만 아직 주식시장은 5%이하입니다. 변화할 여지가 많고 대체거래소의 등장도 이런 변화를 촉진시킬 가능성이 있습니다.
그렇지만 영원한 전략은 없다는 생각입니다. 영원한 비즈니스모델이 없는 것처럼…..
