only four times in 33 years have we lost money at all. and, indeed, if we were to show this another way, 21 of these 33 years we've actually made double-digit returns. if we could go back to the previous chart for a second, you'll see here that compared to a normal distribution, again, the green bell-shaped curve is returns if returns had been normally distributed, you would have expected us to have lost money almost seven times in 33 years, but we've lost money only four of these 33 years. that's good. >> which loss, versus zero or the assumed rate? >> versus zero. >> but assuming 7% average.

quickly on page 12, what we've done here, we now wanted to take a look at the longer term. not the annual returns, but the annualized returns. of course, we don't have enough data, so we did have to use rolling data, and that's using kind of monthly step or monthly rolling. so we have 277 observations. yes, they are related, so we do see the distribution is not normal, but you will see that 60% of the time, the ten-year return was between 5% and 10%. and 31% of the time it was greater than 10%. >> what this also shows is that in all rolling 277 periods of rolling ten-year returns, is we've always earned positive

returns. in other words, compared to the earlier chart where we saw one-year returns, is you'll see returns were a lot more lumpy. and the punch line of that, in the short term markets can be generous or punishing, but in the long term they tend to be more generous and the punishing seems to go away. see how tight this distribution of returns becomes over ten years. in other words, volatility can be very upsetting for very short periods of time, but it tends to go away over long periods of time. >> and i think it's good to look at this is the actual experience. we now reviewed the historical return. what we wanted, what is the expected return going forward? so we took napc's capital market

assumptions and ran monte carlo simulations uses those assumptions to produce a distribution that is with napc short term five to seven-year capital market forecasts and longer term 40-year forecast. so page 14, it's a core slide, it's a busy slide. i'll walk you through, and bill has a lot of insights to share on that, too. so what we've done is i made sure, as i described this, we ran monte carlo simulation for each asset class and for three portfolios. the target allocation, current allocation, and 70/30 allocation, which is 70% equity and represented by mci and 30% bond represented by barclay's ag. we represented the distribution using what's called vox technique, using five numbers.

so we look in the top, you see the fifth percentile, then the first quarter, then you see the median number, third quartile, and then bottom 5% or 9th percentile, so gives you an idea of the dispersion, not just the mean expectations, because this is just one number we wanted to represent the distribution of expected returns or forecasted returns for each asset class. so if you look at the rows, these are the expected distributions with medians that are divided by napc and also the volatility. you will see the largest dispersion is around private equity, lodged dispersion, lodged variability of returns. also the median is fairly high. you will see the medians are

low, but aspersion is also much lower for treasuries and absolute return. so the core takeaway here is you will see the dispersion of our portfolio, the target portfolio, and current portfolio, pretty close, so the expected -- forecasted return and forecasted variability of return is quite close. and much tighter than the 70/30. so we reduced the risk at the result. of the asset allocation. we also see the median is higher than the 70/30. on the upper side, on the fifth percentile, maybe we will underperform slightly. you will see the numbers on the bottom right, where we highlight

the fifth percentile in red. the target allocation is expected to, on the upper right, the really good markets, before 13.6 while the 70/30 is expected to perform at 14%. 14.1%. but look at the lowers, where we will outperform and the first quartile and the worst markets will outperform significantly. >> couple things here. so to summarize this, is we've taken napc's expected return, but those returns have a variability, and it's by asset class, and when you put it together for a portfolio, it was a whole. you'll see here the spread on the far left between the worst and the best, even before a period as long as five to seven

years for public equity, it's a range of 18.6% and private equity it's north of 22%. when you get into asset classes where you expect more volatility and absolute return it's only a spread of 9%, and that's the first asset class that along with class has the most positive expected return, no matter how disappointing returns the markets are. by the way, these are all beta returns. these do not include alpha returns, and i'll comment on that in a moment. if we again look over on the far right side, allen made the comment that you're never going to see a report as good as you saw today, and that is or in the top 1% over the last one, three, and five years. i regret to say that i agree with allen's comment. we are not set up to significantly outperform in a major up market, and you see

that here in the top 5%. okay. our absolute return would be very good. we'd make 13.5% over 7 1/2 years, but we would underperform 70/30 by a little bit. okay, we are not set up to keep pace with a vigorous bull market, because we only have a 31% target allocation to public equity. however, look at the diversified impact of this. even when returns are as high as the 25th percentile, we still outperform 70/30, and look at the positive skew between our portfolio and 70/30 the further you go down the ladder. in the very best case, a market

return, we underperform by 0.5%, a spread of 13.6 versus 14.1, but when returns get kind of disappointing, we do really, really well. so that's what we are set up to do. we are set up to do two things. earn high, long-term returns, and the second is to minimize the impact caused by large market declines. now i said high long-term returns, because these are beta returns, in addition to what you see here for beta returns, we designed this portfolio to earn meaningful alpha returns through manager selection. 1% or north. so -- and so you see that even at 50/50 percentile, 70/30 would deliver you 6.2, according to

nacp, and i actually agree with that. i think mine might be lower than that, but ours is designed to do seven and through good manager selection i hope will do eight. that's a more complete walk through of the five to seven year and the 30 year will be even easier. let's go to the next page. >> and real fast, the way we're set up, we're not going to do necessarily as good as some people might do in a raging bull market. >> right, right. >> that's by design. >> it is by design. >> okay, thank you. >> that is the core -- >> because even if we don't keep up in a vigorous bull market, we're still going to do really, really well. so. >> president stansbury: i'm with you, thank you. >> next page. >> next page, similar analysis with longer term capital market assumptions, and remember what we're trying to do is to give an

idea of the short-term and long-term expected returns, which will be your 50th percentile, but that's just one number. we wanted to show you that we look much deeper. we compose it by asset class and make sure we understand where the diversification is coming from. and so this quantifies the downturn perspective and diversification. moving on, what we've also -- given the capital market assumptions from napc, we also wanted to confirm that we have for the set risk appetite that we discussed, the 12%, our asset allocation is close to the efficient frontier. so we set up the optimization, so efficient frontier is a set

of portfolios that is set up to deliver the highest expected return for the best level of risk. so if we set the target risk at 12%, which is on the x axis, the expected return is for our target allocation is around 7.1%. with a shorter term capital markets, and remember, that's on just the beta, beta parts of the allocation. we're very close, our target allocation is what's highlighting, close to the efficient frontier, so all benefits with the arrow margin in bulk, we are effectively on the efficient frontier. >> the important part of this also is to see how suboptimal 70/30 is, and that's because of the high volatility that is

introduced by having a prominent allocation to public equity. in addition to that, that is exacerbated, that volatility is exacerbated for plans that have cash outflows, which we do, because if markets are down a lot in one year, you xs bait that loss by having cash flows out at the end of that year. then when you have the recovery, say, hopefully in the second year, you have less capital than a strictly passive investment, because you are having recurring cash outflows. there's a difference between time-weighted returns, which is the oft stated 10% return versus dollar-weighted returns. if you had a 50% loss in one

year and 100% gain the next year, you know, your two-year time-weighted return was zero. 100 turned into 50 and 50 turned back into 100, but for $1 weighted, if you are taking out $50 at the end of that first year but first lost 50, then you took it all out in the second year when the market goes up 10% and your dollar weighted return was -100%. there can be a big difference between dollar weighted and time weighted returns, and it is exacerbated by the more volatile portfolio you have in the illustration, the extreme illustration that i just gave. >> but to highlight the target asset allocation, is in our current allocation, presents much more risk at higher returns and higher risk adjusted

returns, and we are at efficient frontier with the assumptions that we received from napc, which are highlighted in the rest of the pages for reference. i'd like to conclude that this is just part of the review of the risks. we are going to present another important part on the liquidity side in april, at the next board meeting. and we'll have a comprehensive risk analysis. >> president stansbury: board members, real quickly, i'll say you've seen a large increase in the volume of the board package and board meetings. i'm going to touch upon real quickly why, because this is one

of them. we're doing so many new things than we did four or five years ago. we're doing private credit. real assets, absolute return, risk management. we have a much more diversified asset allocation. we now have $25 billion in assets. five years ago we had 17, and then lastly is we are emphasizing more unique niche and specialist strategies. so there are really seven core reasons why you've seen the volume of your board package go up. again, we're doing a lot, lot of different things. now, it's showing up in your returns, the numbers that you see allen provided, so the story is good but with the consequence you have a much larger board package and longer board meetings. with that, we'll turn it over to the board.

>> president stansbury: questions from the board? commissioner driscoll? >> commissioner driscoll: two observations. we were doing private credit for a long time. >> president stansbury: not anywhere near -- >> commissioner driscoll: waylaid for different reasons, but doing more now, better, great. other changes in the fixed income market. i guess i'm going to try to articulate on page 14. there's numbers here that suggest that the 60/40 portfolio or the 70/40 portfolio would be a higher return. however, i would say the board's decision is to go for the average return with the highest probability of achieving it. >> that is exactly correct. >> commissioner driscoll: might say it's a more efficient portfolio in one sense. we weigh all the odds, skills, but that's why you might say we're not going for the highest total, but the best odds of making a good return. >> another way of saying it is looking at this, comparing the

third of the left to the third to the right to the last on the right is we have a 75% probability of outperforming 70-30. in other words, we are three times more likely to outperform 70-30 than we are to underperform it, but there's a chance, there's a 25% chance we do underperform, and, again, in addition, these are based only on beta returns. i think that percentage of 75 goes up materially when you include alpha. >> not necessarily the highests return, but the highest probability. >> that is correct. you hit it on the button. >> commissioner driscoll: thank you for letting me say that. >> yeah. >> president stansbury: any other questions from the board? why don't we open it up to public comment? are there any members of the public that would like to address the commission regarding this presentation? >> i noticed that you reduced the assumed rate of return from

7.5 to 7.4. what i can't understand is investing is all about risk and reward, you increased the risk of your investments by buying in hedge funds and now you're reducing awards. i don't understand why you do that. it's all about -- every time you retrieve the risk, low liquidity or anything else, that investment portfolio, i guess you're assuming it's going to get 7.5%. if it gets 7.5% return for the next ten years, ask mr. coker how much money he's going to pay in management and performance fees and consultant fees to get a 7.5 times return. it's simple s&p 500 index funds are going to give you more than 7.5% returns. just take fidelity numbers.

ten-year performance on s&p 500 was 30 -- [ inaudible ] -- you have to get a 16% of return. you have to get at least 2.5% to 3% just to pay the management and performance fees. take that into consideration. never asked mr. coker how much you paid in the last ten years in management and performance fees. ask him on that. portfolio, ten years, how much you would pay in management performances. at best you could probably buy a san francisco office building with the money. thank you. >> president stansbury: i see there are no other members of the public present, i will close public comment. thank you very much. any other questions from the board?

>> request that we address the action items in case we lose a quorum. >> president stansbury: of course. okay, thank you very much for your presentation. let's take a look at what is left in front of us. >> item 11, determination of -- >> president stansbury: why don't we jump a little bit out of order here. let's go to item no. 17, please. >> clerk: action item, review and approval of pension adjustments for melonee alvarez. >> president stansbury: we'll take this as submitted. we'll open it up to public comment. are there any members of the public that would like to address the commission regarding this item? seeing none, we will close public comment. does the board need any discussion on this item? >> so moved. >> president stansbury: the item is moved by commissioner

casciato. do we have a second? seconded by commissioner chu. any discussion? i call for public comment. >> you did. >> president stansbury: i did, thank you. take this item without objection? great, item passes, thank you very much. why don't we, since item no. 17 is now passed, why don't we go back to item no. 10. >> yes. >> clerk: item no. 10, action item. approval of hire recommendation for proxy services. >> sure, board members, i'm going to ask kirk to introduce the item and kirk and andrew have a few comments. >> board members, you may recall that we recently issued an rfp

for proxy voting services. we received two responses, one from iss, our current provider and one from glass lewis, a san francisco-based firm. as outlined in our memo and as andrew will describe further, we conducted due diligence on both firms and are recommending to hire glass lewis for proxy voting services. andrew? >> thank you, kirk. as kirk outlined, we issued this rfp. just as a quick reminder, sfers relies on our proxy provider not only to execute votes at companies where sfers is a shareholder, but also to provide insights to reports and analytics, provide us research on corporate governance and certain environmental and social issues, and to assist in keeping our proxy voting policy current each year as we update it. as kirk outlined, we conducted due diligence on both proxy

voting providers. they are the two leading providers out there, and they formed basically a duopoly with iss having 60% of the market cher and glass lewis having about 40% of the market share. what we really focused on and i'll highlight are four areas. one, does the firm have a stable ownership structure, firm customer service and commitment to its proxy services business, two, can it provide high quality voting services for sfers without error and deliver us compelling analytical reports. three, does the firm avoid conflicts of interest and have strong controls, compliance procedures, and lastly, does the firm offer a competitive fee structure for the services that we require? after we conducted our due diligence, we're recommending that glass lewis be engaged as our proxy voting consultant. we've made this recommendation based on three areas. one, we believe that glass lewis

demonstrates a strong commitment to their proxy voting services business segment. two, we believe glass lewis headquartered here in san francisco can deliver highly responsive customer service to sfers, and, third, glass lewis offered the lowest and most favorable overall fee structure for the services we need. we've discussed the ability to transition services from iss to glass lewis with both firms, and we're confident that that could be accomplished efficiently and along the timelines that we require. >> president stansbury: great, thank you so much for your presentation. are there any questions from the board? there's a motion by commissioner casciato. were you seconding, commissioner bridges? >> commissioner bridges: i will second it, but i had a quick question. >> president stansbury: so we have a second by commissioner bridges. open up for discussion, please. >> commissioner bridges: so besides what you've outlined in

here, were any major failures by iss and the services they provided in the past? i know we've been there a long time. i'm familiar with both firms, nothing against glass lewis, i'm just curious about it. >> no. it was a very competitive process. we've been satisfied with the services iss has provided, and i think as we outlined in the memo, we were confident that either firm could deliver the services we needed. >> commissioner bridges: i agree, and glass lewis is a very good local firm. i would agree with that, as well, so i'm familiar with both firms, just curious, that's all. >> president stansbury: okay, great. why don't we open up for public comment. any members of the public that would like to address the commission regarding this item? seeing none, we'll close public comment. anymore discussion from the board? great, can we take this item without objection? item passes. thank you so much. item 11, please. >> clerk: item 11, action item. recommendation to terminate dfa

us small cap value trust, terminate dfa small company portfolio, and to convert russell 2000 growth to russell 2000 core. >> board members, as you know, we've been reducing our allocation to public equity systemically, and so some well diversified strategies, you know, we've had some actions similar to this, and this is the next. it's in line with reducing our allocation to public equity. >> noted in the title that we're really recommending three actions here. we are -- staff is recommending nepc is in support of the two small cap manages and the russell 2000 growth index, russell 2000 core index.

we've written a brief memo that describes our rationale, and unless napc would like to make comments, we'll turn it back to the board for any questions you may have. >> president stansbury: anything from napc on this? >> no, strategies, some 40-odd, two or three are on our preferred list, but these strategies are not. they have been noted as underperform and have been on your watch list, but the restructuring of the portfolio along the lines kirk described are driving this and we're in support. >> president stansbury: great. we'll open up to the board. any questions for the board? commissioner driscoll? >> commissioner driscoll: glad to help continue this reshaping of the total portfolio, as well as the equity portfolio, particularly, but can you make the case for the conversion of the russell 2000 growth to the 2000 core? not that i'm opposed to it, but where's the data? >> so our u.s. small cap portfolio we have a core

portfolio and the dfa small cap value, and prior to several years ago, we had a u.s. small cap growth manager that were terminating and put into a russell 2000 growth to offset the dfa value. now that we're getting rid of value, we would need to put that into a core just from a portfolio construction standpoint. >> in other words, commissioner, we're terminating one value manager, so we're terminating a growth index so that we retain core exposure. we don't have an overweight to grow. >> commissioner driscoll: okay, i guess i should have figured out how to guess that from the information given to us rather than one table of numbers. then i would have seen it. all i had to do. >> president stansbury: anything else from the board?

>> you want to take these separately or take them -- >> president stansbury: let's take them together. would anyone like to move on all of them together? accept staff recommendation? >> i'll second. >> president stansbury: motion by commissioner chu, second by commissioner casciato. before discussion, let's open it up to public comment. any members of the public that would like to address the commission regarding this item? seeing none, we'll close public comment. any discussion? we take this item without objection? great. item no. 11 passes. where would you like to go from here? do you have a preference? >> we have one budget to continue to the next meeting, we have one additional action item on the gasb report, but jeanette said she wouldn't mind if we just pushed it or took it today. >> president stansbury: what is your preference? >> take it. >> president stansbury: take it? okay.

>> clerk: item no. 15, action item. presentation of the june 30, 2018, gasb 67/68 report. >> good afternoon, commissioners. this is the financial accounting report, and we do this annually. it's under a government accounting standards board rules, and i'll be happy to answer any questions. >> president stansbury: any high points, without going through the whole report, anything you want to point out to the board? >> i would just say that under the board summary, that the net pension liability is $4.3 billion, that is lower than it was a year ago due to our good asset return june 30, 2018. this will, of course, be very volatile, because the gasb net pension liability is based on

market value of assets, so as market value of assets has volatile returns, excuse me, the net pension liability will float with that. >> president stansbury: great, thank you. why don't we open up to public comment. are there any members of the public that would like to address the commission regarding this item? seeing none, we'll close public comment. any discussion from the board? would anyone like to move the item? >> i move acceptance of the report on gasb 67/68. >> second. >> president stansbury: there's a motion by commissioner driscoll, seconded by commissioner bridges. any discussion? can we take this item without objection? great. item no. 15 passes. thank you very much. >> thank you very much. >> president stansbury: mr. huish, do you have a preference where we go from here? >> executive director huish: cio report and finish up the

investment section. >> do we want to do absolute return today? >> executive director huish: do you want to do absolute return? >> president stansbury: i think we should do absolute return today. >> executive director huish: next? >> president stansbury: let's do that one. call item no. 12 please. >> clerk: item 12, discussion item, absolute return portfolio update. >> in december, absolute return was two years and three months old. this is alberto and roberta. i'm not sure if the board has met alberto before. program is now two years and three months old. it's outperformed the hfra index by, i believe, 2% annualized. we're going to ask you to make some introductory comments and walk through the standing of the program. >> thank you, bill, thank you, commissioners. given the lateness of the day, i will attempt to be efficient and stimulating with my comments. i'm going to make a couple

preliminary remarks to set some context on the program. and then i'll walk through the presentation materials that were distributed for the meeting. last saturday marked the ten-year anniversary of the end of a 17-month bear market that lasted from october 2007 to march of 2009, in which stocks globally lost approximately 50% and investment portfolios across the globe lost trillions of dollars. closer to home, sfers' portfolio declined in value by approximately $6.3 billion, or 33% during this time frame, and the plan's funded status declined from a value of 125% to 72%. pension reform followed in 2010, resulting in higher contributions both for the city and employees to elevated levels that today, nine years later, are still the norm. memories have faded, but the

wounds have healed very slowly. in recent years, equity markets enjoyed the longest bull market in history until q4 of 2018, when global stocks declined nearly 13%. calendar year 2018 ended with a repricing of risk assets and a deleveraging as a result of heightened uncertainty in the global macro and geopolitical global environment. slightly two months into 2019, equity markets have once again found their way and have risen over 10%. the question is, is there more certainty in the world to justify this movement and this retreat in value back to the current levels. the answer is unlikely, but the true answer is that we really don't know. the good news is that if the answer to that question is no, in the second quarter of 2019, which we're about to go into,

looks more like the fourth quarter of 2018 or even worse, the fourth quarter of 2008. sfers is in a much better position than we were during the global financial crisis to weather that storm because of the 13% allocation that we have to absolute return. so now i'd like to direct us to the materials. if we can bring up staff's presentation, please. page 1 of the materials here provide a recap of our program objectives and more granular explanation of why sfers has an absolute return program. i'm not going to go through these in detail, and would direct us to page 2, where we provide a high level indication of where we are with this program relative to what was approved by the board and how we got here via the program structure that we have in place with blackstone as our strategic

partner managing a customized portfolio for us that is complemented by our direct investments that are sourced by staff and approved by the board. i'll provide more detail on this shortly. on page 3, you will find a slide that was included in our october 2017 presentation to the board how staff proposed it would implement a 15% allocation to direct return that was approved at the september 2017 board meeting. there are several key milestones that were established at that meeting, and the status comments indicated here, indicate that staff has delivered the build out of this program in accordance with expectations. page 4 provides a summary of the absolute return strategies over the last two years, and the mix of capital among the customized ban portfolio and sfers' direct investment. of note, staff has consistently utilized its strategic

partnership to accelerate the build out of this program and achieve the milestones previously communicated to the board. >> president stansbury: david, can i pause you right there? >> sure. >> president stansbury: all right, i got out of order. i'm ready, i'm sorry. please continue. >> okay to continue on page 5? >> president stansbury: please, please, i'm sorry. >> with respect to the program capital mix, page 5 provides an indication of where we are today and what we are tracking toward with our pipeline of investment activities in a direct part of our program. we anticipate that by the end of 2019, more than half of the capital in our program will be comprised of direct investments sourced by staff and approved by the board. by the end of 2020, we expect this to grow to 65%, the growth of which will be funded by reducing the size of the ban portfolio. so i provided an evaluation here on the evolution of the program. now i'd like to speak to some key performance metrics, so if

we could move on to page 6. here we provide investment returns over various time periods, again, some benchmarks. the most relevant is which is 9 fund-to-fund index. even though this is not an investable index, if an investor were to implement a purely passive multimanager hedge fund portfolio, this is what that portfolio would likely deliver, so think of this as a reasonable proxy for hedge fund beta. we can then use this metric as one measure for calculating the excess return our program has delivered via active approach to constructing the portfolio via varying ways to various absolute return sub strategies and manager selection. the comparisons shown here indicates that we have added 181 basis points of alpha on an annualized basis since the october 2016 inception of the program.

notably, our program has generated alpha during all time periods represented here, except for calendar year 2017, when global equity markets demonstrated exceptional performance and were consistently positive for all 12 months of that calendar year. i'd like to point out that in that environment, our program captured approximately one-third of that positive performance, but then in calendar year 2018, when equity markets turned negative, our program captured only 15% of that decline. this is exactly the type of asymmetry that we are seeking. capture roughly half of the decline when markets are negative, then we capture of the increase when the markets are positive. and i'll come back to that point a little later on. >> president stansbury: when it says 2018, we're talking calendar year 2018, right? >> correct. unless it's indicated there where it's fiscal year 2018. >> president stansbury: okay. >> actually, that's fiscal year 2018-19, just to clarify.

>> president stansbury: so the column on the far left that's titled 2018, what is that column? >> that's calendar year 2018. >> president stansbury: okay, great, thank you. >> on page 7 we show this in a slightly different format but i return back to the rfia benchmark, global equities, fixed income and a more traditional asset allocation. page 8 highlights some relevant risk metrics and comparisons, and here i'd like to highlight the beta versus market and sharp. return portfolio is .19, which is right in line with what we desire. our guidelines are set such that we're targeting an equity beta of .30 or less. note that the hfri beta is significantly higher, but the returns are worse, even though

since the inception of our program, we have been largely in an equity beta fuelled return environment. also note here the very low standard deviation values. we like the fact that the measure is low on a relative basis, but volatility is also a driver of opportunity for absolute return strategies. despite some spikes in volatility over the last two and a half years, we've been operating in an unusually low volatility environment. ideally, we would like to see this number higher on an absolute basis, and this is also consistent with some of the measures that we saw earlier in allen's presentation on volatility at the overall portfolio level and also what ana presented in the risk presentation, where we've modelled out target risk at the portfolio level of 12%, but the realized level of volatility over the last few years has been significantly less than that.

>> david, can i comment on this page real quick? so let's walk through this here real quick. so first, since inception between the absolute return and 70/30, almost the exact same annualized return, but what do you get in exchange for the absolute return program as opposed to 70/30? look at the standard deviation. you have two-thirds less volatility, downsides risk, two-thirds less downside risk. beta is .2 versus .7, and you have more than 2x, almost 2 1/2x risk returns. so very good metrics all around. >> okay. page 9 shows a risk return analysis, and here we highlight that the down capture of our portfolio relative to other return streams, and as commented previously, the down capture is about half the up capture. very favorable and desired

dynamic. compare this to the hfri and the 70/30 portfolios, which have the reverse. they are capturing more of the equity market downside than they are of its upside. also of note is that we produced positive returns in more than 80% of the months since inception. >> president stansbury: can i ask a question, what are we capturing percentage of? of the market benchmark, which is? >> measure here is of the -- >> okay. >> look also this number of up months and down months. 70/30 is 20-7. bonds have actually lost money 12 of the last 27 months, and this program has made money 22 out of 27. >> yeah, and that's a good lead into the next page on page 10, where here we show the same metrics as were on page 9, but

instead a comparison to the bloomberg cap ag instead of global equities. here the story is that our portfolio has a negative down capture, which means that when fixed income markets are losing money, which as bill just commented, they have for about half of the months since october of '16 when our portfolio went live, our portfolio has actually been making money. so that's what that negative down capture metric reflects. moving on to page 11, here we have investment returns for a more discreet period of time that we wanted to highlight. again, given that this is the first significant market dislocation that we have experienced in the last roughly ten years, but also since our absolute return program has been live, and here we highlight that our program generated excess return relative to the hfri. again, think of that as a proxy

for hedge fund beta, but also that during a brief period of time when the equity market had both a significant decline, october through december of 2018, and a sharp bounceback in january 2019, our portfolio is reflective of a more stable return behavior and delivered aggregate performance during the same time period. this demonstrates the nimble dynamic that we're desiring with the absolute return portfolio. it's showing we're protecting value on the way down and we're repositioning to be able to capture return on the way back up. i'd also like to share the results of additional analysis that we routinely complete on our portfolio to measure alpha generation. through our partnership with blackstone, we're provided with access to risk analytics tools that allow us to conduct a factor analysis of our portfolio using more than 5,000 different

factors. we conduct this analysis to determine expected returns based on monthly exposure data that we receive from the portfolios' underlying managers, and this provides us with values that we then compare to actual performance. this analysis has shown that our program is consistently generating alpha relative to what the factor exposure indicates the portfolio should be earning. for example, during the market selloff in q4 of 2018, factor analysis suggested that our portfolio should have earned a return of -4.6%, but the actual return was -3.5%. when the market bounced back in january, factor analysis suggested our portfolio should have earned 2.1%, but the actual return was 2.6%. again, providing us evidence that our portfolio is generating alpha both during declining markets, as well as rising

markets. on page 12, we have a summary of where we are relative to our guidelines and an indication that we are in line with all of those guidelines except the correlation to global equity. one thing that i'll point out here is the correlation is above target, but given that equities have generally been positive for most of the time period measured, this is actually a desirable metric. we would expect that over a longer time horizon and a full market cycle, that this value would be lower. on page 13, we provide investment returns at the sub strategy level. now, recall that we break the universe of absolute return strategies down into eight -- the universe of absolute return strategies down into eight sub strategies and those are defined in the appendix of our materials and we're currently allocating to seven of them. this page shows that since the inception of our program, our

allocations to six of those seven sub strategies have delivered returns that are equal to or better than a relevant benchmark for that sub strategy. it is only the equity sub strategy allocation that has not delivered excess return during all periods and that's primarily due to a more recent focus on lowering the directional equity exposure. this helped us in the second half of 2018, but it was actually a headwind with respect to this sub strategy in the first half of last year. so what does all this mean? this is another measure of whether we're adding alpha in our portfolio, and this is whether we're adding alpha as it pertains to manager selection. the data represented here suggests that we are adding that excess return across the sub strategies. on page 14 -- >> president stansbury: can i ask for a favor? >> yeah. >> president stansbury: for slide 13, i don't see the

numbers in the presentation, and correct me if i'm wrong, but can you send us the numbers for the other comparisons that you used on slide 11, like the barclay's u.s. aggregate, hfri, i guess that's already on there, but the 90-day u.s. bill, can you send us those numbers for calendar year 2018 so we can compare and contrast? of sub strategies? >> president stansbury: yeah. >> we'll send them so it's all on one page. >> president stansbury: just that one page. >> okay. page 14, this is a reflection of our view based on the current market environment and manager opportunities. if we had the ability to snap our fingers today and turn the portfolio into exactly the way we would like it to be allocated, what we would do is have fewer managers in total, lower exposure to equity strategies and multi, and have

more exposure to quantitative strategies, macro, emerging markets, and special situations. so some of the reason of part of the explanation as to why we're not there is that we're still in buildout phase of our program. we're only two and a half years into this, which is relatively early in the life cycle of a program of this nature. the other thing is that the market environment is constantly changing, and the manager opportunity set is also a dynamic one that we're monitoring and acting upon. the desired mix is likely to change over time, but the one thing we feel very confident about is that we will expect that the portfolio will become more concentrated and we will see the total allocation to the number of managers reduce over time. page 15. this is a recap of our

investment committee presentation slide delivered in the summer of 2018, and i'm not going to run through these bullet points. the only comment that i'm going to make is that nothing has changed in our strategy from what was presented last june. >> president stansbury: thank you for putting all of this together. this is something that i've been wanting to see for some time. we needed more data, little more history, i think, and given an opportunity with last year, anything from blackstone? >> yep. >> president stansbury: alberto? >> good afternoon, commissioners. we have our own presentation. go to page 3, please. it's a different document. is it in the same file? >> separate. >> separate? there it is. >> there it is. >> all right, thank you. so page 3 highlights what a tough environment 2017 was for

investing across asset classes. you have a list on the left and right side of different type of asset classes and 2017 was negative for all of them, and this is a nominal basis, but none of these asset classes actually beat inflation. so on a real basis all asset classes in 2017 were negative. this is the first time it happened in many, many years. in decades, actually. so there was really nowhere to hide. bonds, credit, you were negative, and in this context, you know, a well diversified portfolio of hedge funds actually protect the capital pretty well, and after '17, where there were strong returns and as david mentioned hedge funds protected upside and capital in 2018. next page, page 4, let's talk about hedges more specifically, and hedge funds should not really be thought of as just one big asset class, but there was

difference in performance across the different sub strategies, and as david mentioned, you're invested in seven of them and the performance is quite different. when most hedge fund sub strategies were up at the end of the third quarter, you know, the losses in the fourth quarter were directly correlated with the beta of the sub strategies, so long-term equity in terms of a higher beta, that was the sub strategy that lost the most in the first quarter. and equity long/short, driven, credit all lost very meaningfully in the fourth quarter. now, beta was not the only culprit. there was also negative alpha amongst hedge funds. now, on the other hand, sub strategies with lower beta, more training oriented, quantitative, macro, did well and preserved capital. so, you know, when you look at the overall performance in '18 in hedge fund land, there were

clearly sub strategies that sub stra tracted and some that helped. on the next page, shows you at the portfolio level. in 2018, on the right-hand side, the tailwinds, so the sub strategies that helped in performance, were residential mortgages, structure credit ebs, multistrategy and quantitative strategies. these strategies tend to have bigger low equity data and even low corporate credit data, so your typical high yield index type of data. on the other hand, the headwinds came from equity long, short, stress. at the fate of the cycle when the fault rates are low, stress managers tend to have a decent amount of equity, which tend to have a high equity beta, and then you have credit. and so as david mentioned, in terms of area focus for 2019 and beyond, it would be to reduce further the equity allocation,

and by the way, right now the equity allocation for services is 22%, to give you some sense, the overall hedge fund index is almost 40% equity. so the program is already underweight equities, equity long/short, that is, and the goal is to make that even lower. and replacing that would be to add quantitative strategies. this is equity market neutral, quantitative, and adding global macro, and also continue to add to special situations or coinvestments. so if you look at that pie chart, we're reducing the blue and adding stuff on the left-hand side. this is very consistent with the view that ban, this next page, page 6. investment committee meets every month and establishes this type of directive in terms of which sub strategies we feel can deliver more alpha, and we want to add to, and which sub

strategies we want to reduce from in hedge fund land, and you can see on the top mortgages, macro rates, structure credit, emerging markets, are some of the things that we like to add. and you can see at the bottom the ones they want to reduce. namely fundamental equity, event activist. reason being, these strategies have a lot of beta, so they'll run with a 40/50, sometimes a higher net exposure, and our view is equity valuations are full and there is more downside than upside, so the risk of being long equity, long only equity or higher beta to equity is just not there. so within a hedge fund portfolio, we are reducing and proposed to reduce further equity. on the other hand, macro rates, emerging markets, structure credit ebs and we feel the opportunity there is more compelling. now that often sometimes comes with some of these opportunities

of less liquid, so that will come with some liquidity provision capabilities. i mentioned the emerging markets. this is really for us, the way we see emerging markets is on the eos dollar nominated the sovereign debt side, so not the emerging market equities, not necessarily effects, it's more on the debt side and u.s. dollar denominated. >> president stansbury: thank you very much for the report. >> one more page, sorry, and the conclusion, effectively. as we think about '19 and some of the areas of focus, we agree with staff that they should continue to identify managers with which to strike strategic relationships, given your size as a program, and continue to leverage your presence in the marketplace to source coinvestments and the team can be very helpful in this and we

will be bringing coinvestments to sfers and also helping sfers in analyzing coinvestments as they source themselves. as i mentioned earlier about technology and, you know, at bam we are continuing to evolve our technology infrastructure. we brought in last year a new cto, chief technology officer, who is migrating all our systems into the cloud, and ideally we can have a sfers-dedicated ecosystem into the cloud, where you guys can upload all your return streams and then utilize, you know, the analytics tools in a more efficient and powerful way. this is something that's coming, and we'll be very happy to help with. and then lastly, you know, one focus would be to ensure that there are consistent guidelines between one and two, so that as we manage one portfolio, it's more seamless and also allow for

the ability to participate in some longer dated opportunities. liquidity provision and strategies like mortgages and structured credit that would require some relaxation of some of the constraints. so one thing to focus -- to consider, is maybe the board should consider harmonizing some of the guidelines on leverage, potentially even increasing those guidelines and relaxing some of the guidelines on liquidity. >> president stansbury: some of the takeaways here is the programs are a little more than twice as much or right at twice as much in an up market as it has in a down market. bonds are about flat during this period, just a little bit better than that, and have lost money 12 out of 27 months, the program's made 2 out of 27. in 2018, 70/30 did lose 6.5,