Annual General Meeting

00:00:05,600 --> 00:00:46,166

Good afternoon, ladies and gentlemen. If I may just briefly echo David's thanks to Alice as well, as departing Chair of the Company. It's been a pleasure working with you and you've certainly held our feet to the fire as Chair of the Company. So, but I think hopefully, you leave the Company in very good stead and we look forward to the to the next chapter, which I'm delighted to talk to you now and I'm just going to spend a couple of moments talking about the proposals as a whole and how they differ for the strategy in the Company today and I'm going to leave it to Muzo to talk about the systematic investment philosophy and process going forward.

00:00:47,100 --> 00:01:21,833

So just to put some context as to why the changes have been proposed by the Board to shareholders. Principally these have been to address three key things. One is, is the challenged performance that the Company has experienced recently. Costs, they're also very conscious of costs and the desire to provide active management at an affordable, lower cost base going forward and scale also to provide the foundations to better enable the Company to scale going forward.

00:01:21,866 --> 00:01:48,100

And these are three kinds of core principles that they've looked to address. In doing so, we've also sought to ensure that we are making best use of the investment trust structure. So, I'll talk a little bit about the enhanced dividend and the use of gearing, and also best practice today is to ensure that there are sufficient shareholder protections going forward. So, you may have noticed that there are three-year conditional

00:01:48,400 --> 00:02:25,000

performance related tender offers, proposed going forward, as well as the ability for shareholders to tender up to 20% of issued share capital. I think this closes at 1.00 p.m. tomorrow. So just talking through the changes and reflecting on how those vary versus the strategy today. First and foremost, I think, at the core of all of these decisions was to maintain the real reason why shareholders have invested in the strategy in the first place.

00:02:25,000 --> 00:02:59,366

And that is, obviously, the exposure to US equities, and the gentlemen quite rightly pointed out the non-U.S. exposure that we have in the portfolio today. That was a consideration that we've looked at, during this process and the, the new portfolio, will in all likelihood be 100% exposed to US equities. It's going to preserve the value investment style. The benchmark will remain the Russell 1000 Value Index. We think that is an important differentiator versus

00:02:59,366 --> 00:03:27,133

a lot of the other US funds available in the market. The heavy growth style available in the market with the likes of the MAG seven, we think a value investment style, particularly in today's market, provides an important diversifier for shareholders to have access to within the US equity market. And finally, the objective is going to remain the same to provide long-term capital appreciation, but to pay out an attractive level of income.

00:03:28,100 --> 00:03:58,500

And we touched earlier in the Q&A around the the increase of the proposed dividend to 6% annually. It will be paid 1.5% of NAV, per quarter. So that's an enhancement on the current policy, which has been 2 pence per share per quarter, which roughly equates to about a 4% dividend yield. And the final point on versus the existing strategy today is, what was, the BlackRock Sustainable American Income

00:03:58,500 --> 00:04:28,766

Trust with baseline screens and ESG objectives. Again, in reviewing the strategy, we've taken the view that for shareholders, that was not a core reason why they invested in the strategy in the first place. And to better enable the portfolio managers to seek out the best total return by removing any screens and ESG objectives. It effectively opens up the full universe for them to seek the best total returns.

00:04:28,766 --> 00:05:01,066

So that is now being removed going forward. So, what are we doing differently? Well, we're adopting this, systematic active equity approach that's using human expertise. It's using big data technology to provide a much more risk controlled, consistent level of returns that can outperform its reference index. And the team have been running strategies in the US that have a track record of outperformance versus the Russell 1000 Value Index. We're enhancing the dividend distribution, as I mentioned.

00:05:01,200 --> 00:05:30,600

So, making good use of the investment trust structure, as well as the introduction of gearing into the portfolio, which is a live discussion with the Board and the portfolio managers at the moment. And then a couple of points at the bottom there, the management fee. So today the management fee is 70 basis points on net assets. And from tomorrow onwards that would, subject to this being passed, the management fee would reduce to 35 basis points.

00:05:31,533 --> 00:06:05,066

on net assets up to £350 million pounds of net assets and then tiered to 30 basis points thereafter. I should also say that from the 1st of May to the 31st of October, so that six-month period, there will be a management fee holiday for that six-months, so effectively zero per cent management fee. And then finally, just the introduction of conditional tender offers going forward. So, over every three-year period running from the 1st of May 2025,

00:06:05,800 --> 00:06:34,300

there will be a conditional performance related tender offer whereby if the Trust does not outperform its reference index, net of fees, by 50 basis points per annum over that period, then there will be a 100% tender offer available to shareholders at that point in time. 106 And the Board have also stressed their desire and will to grow the strategy. And so, if at the end of that period the strategy

00:06:34,566 --> 00:07:06,433

is not above 125 million pounds in net assets, then they will consider also offering a 100% tender offer. Just one last slide from me which I've covered a lot of these points already. I think the one factor or aspect that I haven't touched on is the number of holdings today. It's a much more concentrated portfolio of around 60 stocks. Going forward, we would expect that to to fall in the range of 150

00:07:06,433 --> 00:07:35,833

to 250, so a much more diversified, risk controlled, basis. But I'm going to hand over to Muzo now to talk about the philosophy of systematic investing. Great. Thanks, Charlie. Thanks everyone for coming today. So, a little bit of a background. This might be quite new to some of you, the systematic approach to investing, maybe asking questions: is this normal? is BlackRock doing something radically different?

00:07:35,833 --> 00:08:02,633

But actually, we're seeing more and more interest in our process and more and more interest in our products across Europe. And I think people in the current climate, in the current environment, understand that in many different areas of the economy, you need to use technology to make humans more productive. So, for us, data availability, this is a key driver. There are so many digital traces of human behavior that we can capture now:

00:08:02,633 --> 00:08:28,700

trading behaviour, corporate behaviour, consumer behaviour. There's a lot we can do that we couldn't do 20 years ago in terms of the data. You don't need to put an analyst outside the Apple Store when they launch an iPhone. You can track web trends to see if this new model is more popular than the last one. Computing power as well. We can just test a lot of things. We can try lots of ideas.

00:08:28,700 --> 00:08:49,366

We can try and test and combine, how should I take this piece of data and this piece of economic data and this piece of online data, blend them together and make the best investment decisions. Again, this is something we didn't have when I joined this team 15 years ago. If I wanted to run a back test to test a signal in the US market, I had to run it overnight.

00:08:49,366 --> 00:09:18,533

Now I can do it in 1.3 seconds. So, we just have more tools and better tools. Then finally, diversification. I think more and more investors are recognising that maybe you do have some concentrated fundamental managers in your portfolio, but combining them with a sort of higher breadth, more diversified, systematic process can be quite nice. Certain markets favour certain investment approaches and if you combine 160 those different investment approaches together in your portfolio,

00:09:18,900 --> 00:09:46,400

it can just generate more consistent outcomes for you. So then team that I'm part of, you may have never heard of us. We've been around a while. We do manage quite a lot of money, so we manage around $250 billion dollars across predominantly long-only strategies, regional and global. We've been around nearly 40 years. So, as I tell a lot of my colleagues, nothing really even starts to reach its potential until it's at least 40 years old.

00:09:47,633 --> 00:10:13,266

And we're quite a big team, over 100 people now as well. Now, what do these numbers matter? How do these numbers help you? Well, some of the things that I've talked about, you can't do those with a small business. We spend around $15 million dollars. The team I'm part of, it's been around $15 million dollars a year just on alternative data sources, online data, consumer related data. We spend a lot of money on computing power.

00:10:14,300 --> 00:10:37,700

We also need people around to test the data. Is this information that will tell us something about stock prices, or is it just noise? Is it useless to us? You just need smart people with a data science background to work together to analyse this data. So, you need lots of people and you frankly need money to do it. So having a large asset base means that we can do these

00:10:37,700 --> 00:11:05,133

sophisticated things, but at the same time pass economies of scale on to investors and cut management fees. And then in terms of the experience, yes, we build models that combine lots of data to better predict stocks, but there are humans in charge of that. Those humans have got a lot of experience of managing these types of models through the financial crisis, the Eurozone crisis, Covid, Russia invading Ukraine and the reemergence of inflation.

00:11:05,133 --> 00:11:31,633

So, there's a lot of sorts of institutional knowledge there, and then there's also a lot of research knowledge. These data sources I've been talking about, if I or someone leaves the building, the data and the research and the models are still there. So, you get that the full benefit of that multi-decade experience. What are we trying to do? We're trying to do what every investor does, which is beat the market,

00:11:31,833 --> 00:11:51,033

but we're trying to do something that's a little bit harder, which is consistently beat the market. And so, what we tend to do is have this more diversified approach where we take lots of small bets. We don't go all in or all out on a particular stock. If a stock has a big benchmark weight, we'll probably hold some of it. If the model doesn't like it, we might be underweight.

00:11:51,033 --> 00:12:18,300

But what we're trying to do is provide investors with an outcome that's similar to the benchmark that we're related to and similar to their expectations. If value stocks have a good or a bad year, this Fund will have a good or a bad year, hopefully with a little bit of a better outcome. But we don't want to provide an outcome where value stocks do really well, but we just held the wrong value

00:12:18,300 --> 00:12:44,233

stocks. So, we take lots of small bets, spread the risk out and try and generate consistent performance. If you hold two stocks and one of those has a bad year, that affects your whole portfolio and blows a hole in your track record. If you spread the risk out across lots of small positions, that reduces your risk and enables you to generate typically more consistent performance. Cost is also really important for us.

00:12:44,533 --> 00:13:07,933

Our models will never do a trade without considering transaction costs. If the model thinks that it can add 1% return to the portfolio, but it's going to cost you 150 basis points of transaction costs to do the round trip of getting in and out of that, the model won't do it. Whereas if there's a situation where there's a small transaction cost to earn a larger return, the model will always do it.

00:13:08,833 --> 00:13:36,766

And then, finally, it comes back to this point of managing risk. We are not going to ignore a sector. We're not going to say, well, value stocks are unexciting, I believe in the whole tech story. This is always going to be small incremental risks. You will always get a value portfolio. but hopefully with our process you'll get a slightly better value portfolio and slightly better returns than the Russell 1000 Value.

00:13:38,733 --> 00:14:03,866

So, as I mentioned, we've been around a long time. What have we been doing? Well, we have our roots actually in value. We started 40 years ago with three signals, price to book, price to earnings and dividend yield. And really you can think of everything since then as just trying to make our value bets smarter, capturing all of the other information that drives stock prices, avoiding falling into those value traps.

00:14:03,866 --> 00:14:31,400

And so, we discovered a lot of the same factors and data sources that other quant investors discovered through the 90s. But really, over the last 10 or 15 years, the innovation has accelerated and it comes back to what I've been talking about. How can we just use data and technology to get ahead of the market? Every day we receive about 6,500 broker reports - a human being can't read 6,500 documents.

00:14:31,633 --> 00:15:07,100

We have algorithms that read them and capture shifts in sentiment, specific topics that are being mentioned. Every day consumers are interacting with brands on social media, people are blogging or talking about stocks, companies are holding earnings calls or company management are being interviewed. There is just every day a massive amount of data out there on corporate, economic and consumer activity and so, we are finding, you know, the most efficient ways to capture that.

00:15:07,100 --> 00:15:35,500

We started off actually in 2007, so nearly 20 years ago, with very simple algorithms that would count positive and negative words in financial documents. The idea here being it's a bullish or a buy signal if the net count of words is more positive than negative. Fast forward to today, some of you might have experimented with tools like GPT. We use similar technologies to make sure that our algorithms and our machines read like a human.

00:15:36,100 --> 00:16:05,400

They read in terms of context and backward looking versus forward looking statements. They understand certain topics are important. We've got a database of 27 million broker reports going back to the 90s. That's brilliant for training models. If you've got 27 million broker reports and you've got the stock returns for all of those stocks, you can then train models to read broker reports and understand where the buy and the sell signals

00:16:05,633 --> 00:16:34,633

are, even in the absence of those brokers actually changing their views. And you can do similar things with new stories and earnings call transcripts. So fundamental managers meet company management, we virtually meet them by eavesdropping on their public conversations with other investors. But again, nothing that we do is weird or wonderful and black box. These are standard things that you would expect an investor to want to do. Just capture what information is out there about companies today

00:16:34,633 --> 00:17:00,900

and then incorporate into your investment views. The other thing that I mentioned, just on the bottom right hand side is trading. We benefit from being part of BlackRock. On any given day now, the systematic team, we cross or match around 30% of our trading entirely within BlackRock. So, if it just so happens that another product in BlackRock, perhaps an index fund, is selling a stock as part of a basket to meet a redemption,

00:17:01,033 --> 00:17:27,500

and our models like that stock and they want to buy it, the BlackRock trading desk can put those trades together, match them. There's no market impact and that feeds through into lower transaction costs for the portfolios. As I mentioned though, there are some very experienced people in the team overseeing these models and these signals I've talked about. So, Raffaele and Jeff our co-CIOs. I think they've been with us 19 and 20 years, respectively.

00:17:27,666 --> 00:17:55,900

Ron Kahn, our head of research, has been with us 25 years. Travis Cooke, head of North America Strategies, 25 years. Dave Piazza, head of emerging markets, 28 years. Simon, head of Europe, 20 years. I've only been here 15 years so that makes me sort of a mid-term in terms of tenure. What do these people do? Well, the research team, it's their job to evaluate data sources, to say, does Instagram data predict stock prices and why?

00:17:56,666 --> 00:18:24,300

In that instance, the researchers wrote a research report that showed statistically that Instagram activity related to a company's brand, actually predicts traffic to that company's website. So, you go from seeing something on Instagram to going to the company's website, browsing a product, or potentially buying it, which then predicts company revenue. So that's what our researchers are doing. They're trying to find data that predicts company fundamentals and predicts stock prices

00:18:24,300 --> 00:18:51,233

and it's the job of the portfolio managers to take those signals, to take those data sources that have been rigorously tested and then combine them together to build models that pick stocks in a particular region. So, it's Travis’ help with other people in the team to build a model that's very good at picking stocks in the US market. The US market is very efficient so certain things have been arbitraged away. You can't trade simple signals in the US.

00:18:51,400 --> 00:19:17,066

But there's also a phenomenal amount of data: online data, trading data, that if you can capture that, hen you can continue to have an edge and get ahead of the market consensus. The job of the portfolio managers is also risk management. That's another thing. We try and build models that capture all of the important information, but sometimes something shocking happens in the world. You get a pandemic, one country invades another.

00:19:17,400 --> 00:19:45,900

In that particular situation, it's the role of the portfolio manager to say, does the model have information? Does it have good data here, or should we just shrink the risk? During Covid, there were situations where there was a little bit of risk control. The model sees, for example, REITs as having a certain level of risk. 345 But if you're a restaurant or hotel or office REIT during Covid, you're experiencing a massive amount more uncertainty and volatility.

00:19:46,200 --> 00:20:10,900

What the portfolio managers here are doing is saying, I'm going to second guess the model. I think we should go full tilt into these stocks, its typically reducing some of the active bets in these to make sure that we're not 351 walking, walking into risky situations blindly. Okay, so I've talked about signals and data and things like that. Let's bring it all together. How do we invest? How do we pick stocks? 356

00:20:11,066 --> 00:20:38,966

Well, everything that we do is trying to answer the same questions that all investors answer. We are just trying to do it with data set. All investors care about fundamentals. Is this company attractively valued? Does it have good growth prospects? What's the profitability outlook? Does it have moats that would enable it to maintain its margins? So fundamental investors try and find ways to evaluate these things. We use data to evaluate them.

00:20:38,966 --> 00:21:11,166

So, for example, moats, we now have algorithms that read broker research and look for mentions of things like competitive advantage, pricing power, brand strength. They can then produce a moat score for a particular company. So again, nothing strange or black box here. It's just using data to capture the potential strength of a company's moats. And so we have various different data sources that give us a feel for the valuation and the growth outlook.

00:21:11,166 --> 00:21:39,833

You can now cost growth as well. You know, I've talked about social media activity. There are web traffic trends, consumer transaction data. We get business to business, electronic invoicing data. You can almost in real time forecast growth in a company's revenues without relying on sort of backward looking or, you know, or broker forecasts for earnings growth. If you want to generate consistent performance though, sometimes fundamentals are a little bit slow.

00:21:40,000 --> 00:22:09,266

A company releases a new product, the management changes. There's a new theme or trend in the market and that's why we try and capture shorter term sentiment to navigate turning points and the outlook for the economy or company. So, we do all of this text analysis on broker reports, on earnings call transcripts on the news. Because when something significant happens, people talk and write about it. And so, we're trying to systematically capture that.

00:22:09,266 --> 00:22:38,400

Are management getting more or less confident? Are brokers subtly showing in some of their language that they're starting to get a little bit more bullish or bearish on a stock before it's been upgraded or downgraded? Capturing these, these short-term shifts in sentiment can help you to generate more consistent performance and again avoid things like value traps. We also track investor sentiment. Looking at who is trading a stock can give you a feel for the short-term outlook.

00:22:38,700 --> 00:23:02,333

Have hedge funds suddenly start to dumping or shorting a stock. That again might tell you that something's changing at the company level. Have certain investors, foreign institutional investors started crowding into a stock which might suggest actually it's a little bit overbought and the long-term value and quality story might be there, but maybe you want to wait because you may be able to pick this stock up a little bit cheaper in the future.

00:23:02,966 --> 00:23:30,566

We also think about macro themes okay. This is a high quality, attractively valued company. But in the current environment do they export or import across borders? Do they generate their revenues in the US or Europe or China? We then have data that looks at where companies are exposed, not just in terms of their revenues, but what is their manufacturing footprint, where are their real expenses and then look at the outlooks for these different economies.

00:23:30,833 --> 00:23:57,300

Again, trying to use data to get a feel for whether the, you know, the outlook for the Chinese versus US versus European economy is more bullish or bearish, and do you want companies, can you identify, companies that are more or less geared to that? And then finally, ESG. This Fund doesn't have ESG objectives, but actually, through a lot of people's focus on ESG over the last decade or so, we found some nice data sources.

00:23:57,300 --> 00:24:28,900

That capture both ESG and fundamentals. So, I'll give you some examples. Companies with good benefits that treat their employees well, tend to have happy employees, higher labour productivity, higher return on equity. Companies that occupy environmentally efficient buildings tend to spend less money on energy and have higher operating margins. Companies with good governance and good cyber defence then don't do controversial things or don't lose their customer data and then lose customers.

00:24:29,200 --> 00:24:54,966

So, there are some of these data sources that we found that look on the surface like ESG, but actually they help you to get ahead of the market in terms of capturing the quality and the profitability of firms. The next slides got a few examples, but I think I flagged a lot of these. Again, it's about this theme of using data and technology to answer the same questions all investors want to answer.

00:24:55,133 --> 00:25:21,000

So, fundamentals – some people will build a spreadsheet and talk to people out there and try and look at different data sources. We just do the now costing. We capture all of this online data and offline data about consumer activity. We read, or our models read, to capture the sentiment because that's what all good investors do. Any fundamental investor will spend hours and hours a day reading. We're just trying to systematically replicate that.

00:25:21,000 --> 00:25:52,266

That and have a virtual team that reads literally everything. Macro themes, it's about gauging the top-down versus bottom-up outlook for companies. This stuff matters. Are you exposed to AI megatrends? Are you exposed to tariffs? Are you exposed to European versus US versus Asian defence spending? It's all about recognising that if there is any information that can move stock prices, you've got to incorporate it into your models, especially if you want consistent performance.

00:25:52,266 --> 00:26:19,566

Sometimes the world is quiet and you can just go away plying your trade, trying to predict companies’ earnings and sometimes the world is a little bit noisy and loud and you need that top-down macro view. Otherwise, your model game will just walk into sort of macro driven value traps. What do you end up with? So, all of these things I've talked about can be captured with a score.

00:26:19,766 --> 00:26:40,533

So, whether you're talking about year on year growth in web traffic, whether you're talking about a financial ratio, whether you're talking about the moat score, these will end up being scores of plus 3 to minus 3. We then, or a model, adds all of these scores up to form one score for a stock and that becomes your return forecast. That then means when you've got the return forecast for a particular stock,

00:26:40,533 --> 00:27:07,900

how much you expect it to outperform or underperform the market, you can take into account the riskiness of that stock and build a portfolio that is expected to have the highest level of active return for a given level of risk. And what you end up with is a very diversified portfolio that looks quite similar to the benchmark, but has small overweights and underweights. Obviously, we've got to outperform somewhere. But you'll see here, for example, in the center of the bars,

00:27:08,433 --> 00:27:37,200

if financials are 23.1% of the Russell 1000 Value at the time that this was snapshotted, the model liked financials, but it was about 60 basis points overweight financials. So again, if the model is wrong on financials in the short or the long term it's not going to blow a hole in the portfolio performance and we've taken lots of other incremental bets across various different sectors, various different stocks. It's all about having that,

00:27:37,200 --> 00:27:58,766

you know the law of large numbers. If your hit rate is above 50% and you apply your model to lots and lots of different bets, that's how you can generate consistent outperformance in the long run without putting investors in a very difficult position of underperforming the market by 5% in a year and then what do you do, do you take that loss, do you sell? We don't want to put you in a situation.

00:28:00,666 --> 00:28:35,133

So, I don’t know if it's worth flagging, we do have some performance data. So, we, the team that I'm part of, does already run in the US mandates that have a benchmark of the Russell 1000 Value. So, we do have a track record of specifically applying our models to this benchmark. And you can see in the bottom table there, net of fees over five years, we have outperformed the benchmark by around 70 basis points.

00:28:35,833 --> 00:29:04,433

So, it's that, that consistent incremental return versus the benchmark. These things add up in the long run. If you can, you know, equity markets tend to return so 8 to 10% in the long run. If you can actually compound at 9 to 11% on a 10, 20, 30, 40 year time horizon, that becomes phenomenally powerful. And that's, that's what we've achieved across many of our strategies and what we aim to achieve going forwards.