Barclays : Q1 2020 Analyst Breakfast Transcript

Investor Relations

Barclays PLC Q1 2020 Results

5 May 2020

Sell-side breakfast Q&A transcript (amended in places to improve accuracy and readability)

Tushar Morzaria, Group Finance Director

In terms of how we'll run this today, I'll make some introductory comments to focus on some of the key points we tried to make on the [earnings] call. Hopefully this is not new news to you, but I thought I'd just be helpful, so you get the emphasis on what we found important.

So I'm going to make five comments. I'm going to cover impairment, capital, income trajectory, particularly on the consumer side, reminder of some of the guidance we gave on CIB top line, and a reminder on some guidance we gave on costs.

On impairment, you'll all be familiar with the break-up of the £2.1bn charge that we took in the first quarter. To go through the building blocks, I think of that as £0.4bn normal impairment charge, before we got into the full throws of the pandemic in the latter stages of the quarter. And then, in addition to that, as we got into the latter stages of the quarter, a further £0.4bn in respect of various single-name wholesale charges, many of which were the immediate casualties of the pandemic itself, so relatively vulnerable credit that we impaired straight away.

And then, in addition to that, in round numbers, a further £1.4bn for what I call the book-up, if you like, and that's running our macroeconomic forecasts for this year and beyond, shocking, obviously, GDP and unemployment. Unemployment's probably the more sensitive parameter for us there, and we took an 8% low point in the UK, expecting that to persist for some time in 2020, and 17% unemployment in the US, again expecting that to persist for some time [in 2020], and some fairly significant reductions in GDP, approximately 50% in the second quarter.

The other thing we did was assume a low oil price. We assumed a $20 WTI oil price, a 50% probability of that happening throughout the whole of 2020. That's still below the future strip as I looked at it the other day. And then, the other thing we put into the mix, obviously we were carrying £150m [of what many people] called a Brexit overlay for UK economic uncertainty, that was just rolled into that scenario as well.

Obviously, there's an enormous amount of assumptions baked in there, and we're barely a month or so into this medical crisis, so I have no doubt in my mind that there will be economic revisions and forecast changes, and we'll adapt our provision accordingly as they come through.

In terms of how the models will work, there are massive caveats on this, because I don't think anybody's brave enough to say they have forecasted the economy perfectly for the next year or two. But if our forecasts turn out to be the real ones, the way impairment will work under IFRS 9 is that £1.4bn for book-up, as I called it, should be a pretty good approximation of additional cash losses [that

1

we would experience over the scenario horizon], over and above the book-up that we're already carrying.

So for us, given that we're expecting a recovery in 2021, we would expect, and I'll just use round numbers to keep it simple, approximately £1.5bn further of cash losses over and above what we would be normally running at. We normally run at somewhere around close to £500m a quarter. You'll see last year's impairment charge was £1.9bn, so again I'll just use round numbers for illustrative purposes, roughly £500m a quarter.

In addition to that, from this point on, we would expect to have another £500m a quarter for the next three quarters, so approximately £1bn for three more quarters. Now, obviously that's hugely caveated, because that's assuming that our forecast happens to be correct and an accurate forecast of the economy. We'll revise accordingly.

And the other thing, of course, is I think when everybody was writing their IFRS 9 models, no one had this kind of immense shock and indeed a recovery. There was no empirical data to calibrate these models, so there is definitely some chance that the models [or forecasts] don't capture the real experience correctly. We'll see as that rolls through in the next three quarters, but hopefully that gives you a sense of how IFRS 9 works.

As we get into the recovery phase, into the early part of next year, that book-up ought to be released if the economy is growing, forecasted unemployment levels start coming down, and GDP starts picking up. So you'll get this slightly weird effect where we've take a book-up, we'll then book cash losses, so you'll get almost a doubling up effect, and then you release the book-up almost when you're back into recovery. So that's the procyclical nature of IFRS 9, doubling up into the trough and then probably releasing the book-up as you go past the trough and into a recovery phase.

The other thing I'd say […] is we published our coverage ratios for the various loan types, and one of the things that we pay quite a bit of attention to is looking at our coverage ratio for stage two [loans], particularly in unsecured lending. You'll see from our disclosures in our results analysis that this was 22.6%. That's across all of stage two, so those that are measured as delinquent, some sort of past-due payment history there, or even those that are aren't past-due at all, and we split that out as well.

Across total unsecured lending, so rolling in stage one [loans] as well, that gives us a coverage ratio of a little over 10%. It remains to be seen whether that feels right or not, but that was an important focus for us, as we were just making sure that we calibrated our scenario as appropriately as we ought to.

In terms of capital, we obviously had some significant headwinds in the first quarter. The impairment charge itself took just under 70 basis points off our ratio, and then we had fairly significant RWA procyclicality. As I look into Q2, it's probably a bit early to be guessing on impairment, we've only really finished April, and I think as we go through the lockdowns or releases of lockdowns, economists will start revising their forecasts accordingly. So we'll see where we go there, but away from that, there will be some further procyclicality in RWAs that we will pick up in the second quarter, so I do expect our ratio to go below 13%.

If that turns out [to be] the end of the extreme stresses we've seen in the market environment, and we've seen the low point of economic behaviour and market stress, and there's [some form] of recovery, then I think we would expect to see capital build from that point on in a more regular way, the RWA inflation very slowly reversing, but also the bank continuing to be profitable in that scenario.

2

Two other things I'd just call out, one is pillar 2A. We would expect some level of recalibration during the next cycle, that usually comes through in the second half of the year, either late Q3 or early Q4 if the regulators stick to the normal timetable, which I think is likely in this case. There may be some recalibration there that ought to be of benefit to us, so that will lower our MDA. With that in mind, and also just looking out as best as we can forecast at the moment, I'd expect us to be comfortably above our MDA level from everything we can see now. I don't think that's something that will be of particular concern, at least to management here.

Let me move on to income. We gave some very specific guidance on consumer income, particularly in BUK. Just to remind you, there was £250m affecting 2020 from the lower rate environment, a further £150m from the removal of certain fees and lower overdraft balances, and a further £100m from additional actions that we will be taking [to support customers]. In addition to that, things to be aware of, a typical activity for us is to conduct debt sales on our unsecured book. We didn't do any [of note] in Q1. It remains to be seen whether we do any for the remainder of the year, but we'll be driven by market environment for that. I think in this kind of choppy markets, if it stays like this, it's unlikely we'll do much.

The other thing is that [we published] spend balances on our slide, and you can see that we had approximately a 50% reduction in cards spend as we were getting towards the back-end of March. That obviously means that there's going to be much lower interest-earning balances in our cards book, and that has continued into April. There does seem to be an interesting customers deleveraging, as we can see it. Rather than using their credit cards as a liquidity line, balances continue to contract, as they were, alongside spending being lower as well. So that'll put some further pressure on income, depending on how that plays out.

With all that in mind, when we put that into the mix and we look at NIM for the rest of the year, we expect it to be somewhere between 250 and 260 basis points, again subject to the usual caveats. We're forecasting in a particularly uncertain time at the moment. I expect the low point of that to be in Q2, and that's really because of the [time lag on] deposits that we repriced. You'll be familiar with the lag in repricing deposits, just the usual process that we have to go through, the FCA rules of writing to customers with plenty of notice. So those deposits won't reprice until July, so we'll only get the effect of better liability pricing in the second half of the year.

I'll just also state, we didn't spend as much time on this on the [earnings] call, but even in CC&P you've got some meaningful income headwinds there as well. Very similar to BUK in the sense that cards spend in the US, at least on our data but it seems very consistent with the industry data, is down [by more than] 40%, and balances are again declining there, as they are in the published data in the US. There's a natural headwind there.

Of course, if the economy starts to recover [and] spend levels increase, then I would expect the balances to respond at least somewhat accordingly [both in the UK and the US], but that remains to be seen. On merchant acquiring, which is also included in CC&P, obviously interchange fees are going to be materially lower as a result of the lower spending behaviour with consumers.

The other final thing I'll just state on the consumer businesses, which may be of interest and help, is delinquencies. We have seen a pick-up in April delinquencies. Not very sharply, but it has picked up. As we looked at the back-end of April, there was a levelling off, so it wasn't a continuing climb as we went

3

through the end of April. That may be the result of government support actions. Again, it remains to be seen, but I thought you might find that interesting.

Moving onto CIB, obviously we had a very decent top line in the first quarter. We did call out that April Markets Sales and Trading revenues are running well ahead of the run rate that we had during the second quarter of last year, and you'll all be familiar that it was our strongest trading quarter for 2019.

Again, big caveat here, it remains to be seen what May and June are going to be like, and indeed for the rest of the year, but April at least started off well. It's somewhat important for us, because it allows the income headwinds on the consumer businesses to be at least partially counter-weighed [by better income performance in the CIB], as you saw in the first quarter. So we'll see how we go from there, but April seems ok.

Finally, on costs, you'll recall Jes guided to broadly flat [costs] for FY20. There are lots of plusses and minuses in there. For example, we have decent headwinds because we're ramping up in our collections department, so that will add costs. And doing these kinds of things in a work-from-home environment is a pretty clunky way of doing it, so it's not the most productive way of doing it. It is inefficient, and therefore more costly to be doing these activities from home.

We've also put together a £100m community aid package and suspended various restructuring programmes. We won't do any lay-offs for the immediate moment, so that in itself will have an inflationary effect to our cost line, against which, of course, [we have lower] travel, entertainment, and real estate costs. We're not operating our main offices, so the maintenance costs of those offices are dramatically lower, and various other things. So I think there are puts and takes there, and it's subject to FX [being] broadly flat, and obviously we'll see what the FX rate is for the rest of the year.

Within there, variable compensation is really a decision for the end of the year. Just to remind you that we did accrue an appropriate level of compensation in Q1, and those decisions are being made as we go through the year. That'll be another variable factor in that broad guidance that Jes gave, and obviously that'll track income accordingly.

So I will pause there and open the lines up.

Rohith Chandra-Rajan, Bank of America

A couple [of questions] on net interest income. Firstly, on the margin guidance, and particularly on the recovery post-Q2, you talked about the deposit pricing changes coming in at the beginning of July, what else are you assuming on the pace of recovery in terms of the reinstatement of overdraft interest fees?

In terms of the rates impact, the deposit side comes in July, but I guess the asset side is a bit more immediate, and then there's a hedging impact as well, which is more evenly phased. [What is the] interplay of all of those dynamics in the pace of margin recovery for the second half of the year?

And then the second [question] on loan volumes. You've been very clear around some of the pressure on the cards businesses, I was just wondering what you were seeing in the mortgage book in the UK. You obviously had a strong Q1, one of your peers has talked about volumes down 50% in April versus the Q1 run rate in terms of new lendings, so I was just wondering what you were seeing there as well?

Tushar Morzaria, Group Finance Director

4

On NII and the pace of recovery, I wouldn't say it's a super-sharp recovery, but we are assuming that some of the temporary actions we have on things like overdraft fees [will] reverse at the second half of the year.

In terms of the broader rates impact, so talking here about where asset margins and structural hedges go. The structural hedges, as you pointed out, will just roll through and are relatively straightforward to model. They're not hugely sensitive in a short period of time if you're just looking at this year's NIM, and even if there's a rate back-up in the second half of the year, because it grinds in over a period of time. I don't think it'll make a huge difference to this year's NIM, much more affecting the following year's NIM and NII.

On the asset side, mortgage margins or asset margins have held up ok. Deposits are hitting the zero bound for a lot of banks, [so my guess would be that] there'll be more discipline on asset margins. But we're not operating in a regular environment in terms of regular volumes of business, so again that's a little bit harder to forecast until we get into that sort of environment again.

In terms of mortgage volumes, yes, there's no doubt volumes are lower. A big part of our business has been the remortgage market. There will be remortgages that will continue, so that will be helpful to us. Volumes will no doubt be down from what we anticipated at the beginning of the year, but somewhat cushioned by the fact that our remortgage business will still have a natural flow.

On the positive side, although volumes will be down, our churn margin on new mortgages is positive. So that's, at the margin, helpful, but there'll be muted volumes coming through. But we would still expect to see our mortgage book net grow over the year […] from Q2 onwards, and that's really coming from that remortgage business, [which] we expect to be of some benefit to us.

Jonathan Pierce, Numis

First, a two-part [question] on capital. The PVA deduction didn't move up maybe as much as I would have thought in the first quarter, is there a risk of further follow-through in the second and third quarters? Or are some of the EBA-type moves, which I guess may be applied in the UK on aggregation factors, going to keep that number broadly where it is at the moment?

Also on capital, the guidance for counterparty credit risk add-ons next year, is that still in the same ballpark, despite obviously the big pick-up in counterparty risk-weighted assets in the first quarter?

Tushar Morzaria, Group Finance Director

In terms of the PVA, I think to the extent that there is upwards pressure on PVAs, the cycling through from quarter to quarter, that'll be mostly mitigated on the assumption that the changes that the EBA has suggested putting through are enacted in the second quarter. I think there's a reasonable probability that we'll keep our PVA broadly where it is at the moment, subject to the broad-level market environment of course.

On counterparty credit risk add-ons, at this stage I wouldn't revise guidance. Looking into next year, guidance will be driven by, as we get closer to that time, what sort of market environment we're in. At the moment I wouldn't expect to be revising guidance, but we'll keep you updated as and when.

Jonathan Pierce, Numis

5

On impairments, it's interesting the different ways to think about impairments as the year goes on. Your £500m plus £500m a quarter of potential write-off rate if the models are right would obviously suggest quite a sharp pick-up in actual observed default rates.

My slight concern is that, if that were to happen, the way these IFRS 9 models work presumably would lead to a period where stage one and two provisions actually go up, even though in effect you're releasing from this book-up in Q1. The presumption, otherwise, would be that, as the observed PD is rising, the provisions you're holding against your performing assets and the PD implicit within those performing assets are falling. So is there not a risk here that as the year goes on and observed defaults rise, you also get a further increase in stage one and two [impairment]? Hence we may be looking at an impairment charge that is somewhat bigger than what consensus has currently got for FY20?

Tushar Morzaria, Group Finance Director

I think the way the conveyor belt works is, if the models perfectly forecast the future, then it should have picked up all of the expected losses principally coming out of stage two into stage three. And we do that across the entire stage two portfolio. Not every stage two name will default of course, there'll be an enormous number of false positives in there.

So in principle, in the models, that will be a natural offset. Not everything will jump into stage three and get fully impaired. Some will flow back over time as well, but it remains to be seen. I think it's a fair point that the models were never constructed with these kinds of quite sharp moves in mind, and we'll see where real experience is.

I think the other thing that is interesting, as we look at the data, I talked about delinquencies in our unsecured book, and most of the actions will be in our unsecured book. Delinquencies are up in April, but not by that much. So if you look at our UK cards book, delinquencies are up about 10 basis points in April and began to plateau at the end of April. It could well be that the effect of these government support measures is either delaying or mitigating that fully. It could be that, with the government support schemes and payment holidays, we'll have to recalibrate what the future may look like in a way that's very hard to anticipate now.

It's a fair point you raise, but there are definitely different effects going on that I don't think the models were ever designed to capture fully, for example these government support schemes. If they really are dampening down on delinquencies at the moment, the models were never designed to take that into account, so that'll be another feature that we'll just have to course-correct as we go along.

Joseph Dickerson, Jefferies

On the repayment activity that you referenced, also if you look at Friday's money and credit data out of the Bank of England, it shows a high degree of pay-downs on the consumer side. When you were putting together your allowance, what type of behaviour did you assume?

I would have assumed, for instance, consumers actually draw down, that they don't repay, so I guess I'm just trying to square the circle. If there's this ongoing trend of repayments, what's the consequence on the allowance?

Tushar Morzaria, Group Finance Director

You're absolutely right, every empirical, historical observation that we have on points of stress was not accompanied by a fairly sharp decline in balances. They were always [based on] balances where they were, or indeed picking up, so I think it's back to the comment I made in the previous question. It remains

6

to be seen whether this is temporary. Spending levels have dramatically declined, balances are declining alongside that, obviously not as dramatically as spend levels, but they are trending down. That is definitely helpful in the short term, and you can see that in our delinquency stats where delinquencies probably aren't up as much as most people would have assumed they may have been.

Whether that's a permanent state of affairs and the government support schemes actually work well and we just go back to a more normal period beyond that, or whether when we get to the end of payment holidays or these furlough schemes, that that's when balances start building and delinquencies start picking up is very difficult to forecast.

I think you're right to point out that it's hard to know, but in our models, we didn't assume a permanently declining level of balances. We tried to be conservative in this, knowing full well we don't really know which path we'll be going down, and we don't really know whether we've forecasted the economy correctly.

To the extent the economy is worse and balances are lower, there are probably some offsetting effects there, but I think we'll know a lot more in Q2 as the lockdown is released in most countries. In some ways the UK has the benefit of some European countries going first, so we'll get a better sense of consumer behaviour coming out of lockdown, [whether] spend will really increase or will that be accompanied by balance increases. I suspect releases of lockdowns might start taking effect in the June and beyond time frame, so by the time we're reporting towards the end of July, we may have a couple of weeks of information there.

I think the other next big check point will be the end of the government schemes, which ought to be, in theory, accompanied by the government feeling that they've done their job and bridged us to the other side of recovery. Whether that will be accompanied by a pick-up in delinquencies, or indeed governments have been successful and we're back to a more normal-level operating environment, that may not happen until much later in the year possibly. I don't think I've answered your directly, but hopefully it gives you a little bit more colour.

Joseph Dickerson, Jefferies

Indirectly was good enough. It's just interesting that the government schemes seem to have had the opposite effect on behaviour that one might have thought. On the merchant acquiring side, you talked about the merchant acquiring numbers being down, is there an impact of merchant mix there as well?

Tushar Morzaria, Group Finance Director

Yes, there is. In fact, the published numbers we gave, I think we were down 9% or something like that. We actually have a relatively good market share of acquiring for grocery stores. Obviously, volumes are up there, and high-street-type shops have virtually collapsed to zero, so there is definitely a mix there.

We are probably better represented in the market with larger companies, and those are thinner-margin businesses, and probably underrepresented in smaller businesses where margins are bigger. That may be, at the margin, a little bit helpful to us. They are probably more resilient businesses, depending on which sector you're in, like grocery stores are pretty buoyant at the moment, and most smaller retailers are obviously finding life pretty difficult, so it's probably at the margin a benefit.

The other thing that may be helpful for folks on this call is that spend levels have declined quite sharply on the data we gave. They seemed to be flat-lining again, a little bit like delinquencies, they're not continuing in an incessant march down to zero. Again, it remains to be seen what May looks like, but it looks to be plateauing as we were getting late into April in terms of the decline in spending.

7

Guy Stebbings, Exane BNP Paribas

The first one is on the bounce-back loans, which looked to have been very successful in the short period since they've been launched and, on the face of it, quite a nice little earner given the c.2.5% interest rate, zero cost of risk, and very limited capital allocation.

But I guess the missing part, beyond the fee for the scheme, is then any operating costs. I'm conscious that while the credit risk sits with the government, the banks still do all the heavy lifting in terms of trying to collect when the borrowers default. I'm just trying to work out how much of that headline customer rate you expect to be eaten up, given those costs? And I guess it's similar on the other schemes as well.

And then, the second question I have is just on pillar 2A, to help us gauge the possible move later this year. Appreciate there'll be other things going on in there, but with regards to the RWA move alone, what would be the starting RWA point that we should be looking at? Is it Q3 2019 to the move at H1 2020, for instance, for us to understand how much you then see in terms of the pillar 2A reduction because of just the RWA move alone?

Tushar Morzaria, Group Finance Director

On the bounce-back loans and various other schemes, the bounce-back loans themselves actually aren't at all costly for us to administer in terms of the creation of the loans. These are pre-existing customers and those KYC/AML checks is a pretty straightforward process. I forgot the exact number, but it was tens of thousands that we went through just yesterday alone. It will be several hundred thousands, I guess, in our customer base that are eligible, and those customers can literally go online and it's a super fast process, and relatively straight through from our perspective.

The CBILS loans are much more clunky to administer. We have compliance requirements under the Consumer Credit Act and various other pre-existing regulations, so they are more clunky. And that was some of the frustration at the earlier launch of those schemes - why was it that the banks were not able to turn the taps on very quickly? All the bankers were obviously working at home, a lot of the branches were closed, and small businesses were all working at home as well, so that underwriting process for a slightly more complicated product was clunky. I wouldn't say it was at origination, we haven't been adding bankers or anything like that, so I'm not sure it's cost inflationary in and of itself, but certainly clunky.

At the back end, I think your point is right. The bounce-back loans, I think, are six-year terms, so there's no doubt there will be some folks that will not be able to repay these loans. There's quite a meaningful length of time before that becomes an issue, and that's such a big judgement on what these businesses are going to be like a few years down the line. So probably not an immediate concern in terms of costs, but certainly an issue for further down the line.

I think the real cost for us on collections will be away from government schemes, just unsecured credit and to the extent mortgages or [consumer loans] gets into trouble. That will be [cost inflationary] for us.

On pillar 2A, it's hard to give very specific guidance, because some of pillar 2A is referenced as a percentage of RWAs in the way regulators give the add-ons and some are fixed-pound amounts. For example, the pension add-on is a fixed-pound amount. It gets a little bit murky to put all of that into the mix.

The thing I'd probably direct you towards, and the IR team can give you the link if you don't have it already, is the Bank of England's stress test itself will give you at least a view into how they've thought about it for

8

their own cyclical stress testing. You can look at that over a number of years to see what they've done on pillar 2A.

It's a modest reduction on their numbers. Obviously this is a very different situation and some of the add- ons will behave differently because this stress is so different to the Bank of England's stress, but I think it'll give you at least a starting point, just to see what benefits there may be out there.

Andrew Coombs, Citi

You said that it's important to look at the stage two unsecured, I think you said 22.6% covered. In our discussions with your peers, one of them pointed out that that's only relevant based upon your write- downs time frame and methodology as well, in that if you tend to write off later, you will always have a bigger ECL balance. So perhaps you could just elaborate on what's actually required for you to write down the position, and how you think you compare to peers?

Tushar Morzaria, Group Finance Director

I don't know the answer to how we compare to peers, but the write-downs come out of stage three, so I think that's a more relevant observation. On stage three, our coverage ratio is, from memory, [more than] 70%, 73% or 74%, for unsecured lending. So in some ways, the way we think about it is when something goes from stage two to stage three, we're quote "fully impaired" or at least going to a loss-given default- type level.

When that then gets written off, in principle there ought not to be much of a P&L impact, as we should have got all the way down to recovery value anyway, but it's just a transfer out of stage three and off our balance sheet. I think the stage two is a meaningful comparison, because these are loans that aren't really impaired. These are credits that we don't think are fully impaired yet, but how much have we provisioned across the entire portfolio stage two, and this is where I think IFRS 9 can be a little bit hard to follow. Whenever people have complained about IFRS 9 as an accounting standard, one of the complaints is the number of false positives that it forces you to book.

Historically, […] 80% to 90% are false positives. I don't know what will be the case in this particular scenario, but we are putting [more than] 20% coverage across an entire [stage 2] population, where only a smaller percentage of that population may end up in stage three over the cycle. Again, in this particular medical crisis, it's very hard to forecast what that is, but there will almost certainly be some false positive in there, meaningful false positives. So I think it sort of evens out in that respect.

Andrew Coombs, Citi

To commend you on your disclosure, you do provide a breakdown of stage one, stage two, stage three by product at the Q1 stage. Sadly, not many of your peers have done. We'll have to wait for the interims for them. Just coming back on the write-downs, and your point on stage two I fully respect, for stage three to write-down, what's the time frame and what are the key determinants there?

Tushar Morzaria, Group Finance Director

It's not mechanical, it'll be a combination of the passage of time and, quite frankly, the direct conversation we're having with the debtor in this case. We may be able to put them on various payment plans and [able] to bridge them back into a better situation, or we may not. Or we may even, as I mentioned in the past, put them into a debt sale programme and pass them on to a specialist agency that are much better at dealing with those situations. So I'm not sure there's a straight answer I can give you on that, unfortunately. It's not automatic. People will stay in stage three for some time, and it's rare, but some

9

people do cure back from stage three to stage two. Rare, but it can happen, depending on the specific circumstances.

Robin Down, HSBC

The way I was thinking about the consumer credit book is that you've obviously set out your macro scenario. You're going to have a series of loans that end up defaulting, so stage one, stage two transfers into stage three, but at the same time, you've got this big balance reduction coming through.

I'm trying to work out in my mind how you've allowed for that within both your provisioning and your guidance on the £800m to £1bn going forward in Q2, Q3, Q4. Because with the rate at which the balances are falling, if you don't do anything else, I would have thought the stage one and stage two coverage ratios would actually rise at this point, so you would normally be releasing some of that, assuming your economic scenario persisted.

So I'm just trying to work out what you've actually got in your guidance for Q2 to Q4, whether you've allowed for, in effect, the possibility of stage one and stage two being released from that consumer credit book as balances fall.

The second question was just coming back to the subject of dividends. You've given us guidance of the CET1 ratio dipping [below] 13% in Q2, and then hopefully flat, maybe modestly improving. It doesn't feel like you're going to be back at your CET1 ratio target at year-end. I'm just wondering how you think about the prioritisation of getting back to that target versus resuming an ordinary dividend, if the regulators allow you to, at year-end.

Tushar Morzaria, Group Finance Director

On the first question, […] we've not tried to be too clever there. We don't know with certainty that the economy we've forecast [will play out], and all we've done is just picked the mid-range of publishing economists. We know almost certainly that won't be the case. So there'll be some true-up as a consequence of that, up or down, we'll see.

The ability to forecast where balances are going, again, will just be very hard for us to do. We felt it's conservative not to try and be too clever and assume a permanent declining level of balances. So to the extent that really turns out to be the case, balances don't recover, that'll be something we would revise accordingly.

As I said, I do think we'll learn a lot about this in Q2 and Q3. As and when the lockdowns are released, if we don't see a material increase in spending and therefore a natural improvement in balances, then I guess what's going to happen is the economy's probably not functioning as well as the economic forecasts. We may need to update our macro-economic variables again, and the book-up will be revised. Against which, we'll probably be doing that off a different starting point of balances, and we'll have to put all that into the mix, but I think we'll know a lot more about this in Q2 and Q3. We haven't tried to be super clever on balances for the first quarter.

And remember, we were doing this in the second and third week in April, very early on. Even in the last three weeks since then, we've learnt a lot more about the delinquency patterns that we're currently seeing. And in two and three weeks' time, we'll have a lot more. It's a giant extrapolation off a very small data set, unfortunately. That's what all of the banks have to do, and we'll learn a lot more over the next few weeks.

Robin Down, HSBC

10

But the main point is you've not assumed any kind of releases related to balance functions?

Tushar Morzaria, Group Finance Director

No, we haven't. We just don't know whether that's the case. We will find out.

On dividends, it's a very tough question to answer here, because we're so far away from the end of 2020. In theory, if we're through the medical crisis and we're on a path to recovery and we see our clarity on capital trajectory, then that'll be helpful in terms of determining dividend/distributions back to shareholders. I'm not sure there's much more I can comment on that at this stage. I think it's one probably not even for the interims, probably one for Q3 and beyond, I think.

Chris Cant, Autonomous

I just wanted to come back on ECL. If I can just recap some of the comments you made earlier in the call, you've taken the £2bn at 1Q and you're pointing to £1bn per quarter for the rest of the year. Please correct me if I'm wrong, but I think you said that your macro scenario, if you're bang on, then the £1.4bn book- up you took in 1Q will then start to reverse into 2021.

So if that £1.4bn was reversing in 2021 and you have the regular way £0.5bn or so that you've been talking about, you would appear to be pointing to about £5bn of loss this year and then, potentially, sub- £1bn next year, if your macro is bang on. Is that the correct interpretation of your comments?

And then just in terms of how you're talking about this crisis playing out from a bank loss perspective, and I know you've described it as a medical crisis a couple of times. Obviously there has been quite a lot of nervousness on macro, both from financial policymakers and politicians, but it doesn't seem that you expect this to flow through much into loan losses. Is your view, in effect, that the losses from this crisis are going to be socialised by the fiscal response? Or is your view that actually the macro is simply not as bad as some commentators suggest?

Tushar Morzaria, Group Finance Director

On your first question, in principle, yes, you've interpreted my comments correctly. You get this peculiar effect of, and please don't quote me on this, but over-impairing, if you like. In the time into the trough, you're doing a pre-stressbook-up and then adding onto that the actual cash losses that the book-up would imply, and you get that doubling effect as you go into the trough. Then as you come out of the trough, your cash losses start decreasing at the same time you're releasing your book-up, so you're sort of under-impairing, if you like, on your book-up, and that's the procyclical nature of the accounting, which is just what it is.

When exactly you get that book-up release, I loosely said as you get into the recovery into next year. It'll be as and when economic data effectively outperforms the forecast that we currently have. So when economists start revising to the positive, that's when the book-up gets released. Vice versa, if economists haven't forecast the trough correctly and we do go deeper, then we'll be increasing our book-up.

Your second question, again we've not tried to be super smart here. Economists are, in theory, picking up at least macroeconomic effects of where they see GDP and unemployment levels. They're doing their best to capture the effect of government schemes as well, and of course government will, as we've seen with the bounce-back loans, [roll out] new schemes that I'm sure will be put in place, adapted, revised, and enhanced. Government policy will change, I'm sure, as we go through this.

I'm trying to be agnostic as to whether governments feel that they want to socialise all the losses through government policy or whether the banks should pick up the losses. We're just going off the economic

11

forecasts and running them through the models, but you've raised a good point. The models were never designed, unfortunately, with this scenario in mind. We have no empirical data to calibrate on the effectiveness of all of these different policy measures that we've never seen applied.

And maybe they'll be more effective than we've assumed, or maybe they'll be less effective than what's implied by the economics generated out of economists. I think, again, we'll know a lot more over the next several weeks as to just how this is all going to shake out.

Chris Cant, Autonomous

In terms of the quantum of losses then, if I think back to the pre-crisis consensus, consensus had in about £4.7bn of loan losses over 2020 and 2021. From the discussion we've just had, basically if your macro is bang on and if we do have a sharp V, which means that, going into 2021, you basically release this book- up. We're saying that in a pre-COVID world, we've gone from £4.7bn of losses over the two years, which would have been consensus, to maybe £5.7bn?

It feels odd to me that regulators are throwing so many measures at the wall to ease bank capital, and policymakers are throwing so much money at the situation if the end result for a major bank is about £1bn of additional loan losses over a two-year period, relative to what was expected pre-COVID. Maybe consensus was wildly wrong, but it does feel a little odd in terms of quantum. Is there something I'm really misunderstanding about the nature of this crisis?

Tushar Morzaria, Group Finance Director

I can't comment on previous consensus, but last year was less than £2bn of impairment, but your point's still a good question. The way I think of it is if governments weren't doing what they were doing. At the moment, we've got a very deep V shape going on, it's an instantaneous shock and a relatively quick recovery, albeit […] the size of the economy takes a while to get back to pre-COVID levels. But the fact you're in recovery and the fact that unemployment starts declining is the most important thing in terms of how our credit books will perform.

You've got the weirdest situation, where I think in April where you went from, certainly in the US, record levels of low unemployment to record levels of high unemployment within the space of three or four weeks. And then, at the moment, economists are suggesting these record levels of high unemployment in the next few weeks are going to start going back down again.

I think that's entirely a function of that second bit of government policy. If government policies weren't in place, then maybe those unemployment levels will continue to rise or indeed persist. And then I think you'd get a very different shape of credit losses on the banks. The longer you delay the recovery, I think the pain gets amplified quite materially onto the banks, and I think that's what government policy is partially aiming to do, to bridge as many people to the recovery phase.

Chris Cant, Autonomous

I'm surprised at the quantum of the numbers. It feels like we're probably looking at the wrong thing, worrying about your ECL disclosures and stage one and two coverage, when we should be more worried about the income loss, just in terms of the quantum of numbers I said before, but I'll leave it there.

Martin Leitgeib, Goldman Sachs

In terms of credit underwriting, if you look at the UK, has anything changed in terms of how you underwrite, whether that's mortgages, credit cards or general loans, over the last few months? I just note that a couple of weeks ago it seems like a couple of banks have withdrawn 80% loan-to-value mortgages,

12

but that seems to be more driven by operational constraint rather than risk appetite. I just wanted to check if anything has changed in terms of your risk appetite.

Secondly, related to that, you disclosed that 87,000 UK cards are currently under repayment holidays. Could you just elaborate how the process works from here? Will you have to contact them within the next three months and how will they get assessed? How worried are you about a portion of those essentially having payment issues?

Tushar Morzaria, Group Finance Director

On underwriting, I think essentially a lot of the people that need financing now, the government schemes are designed to help them out, rather than originating new positions on bank balance sheets. So I'd say there's probably a fairly high degree of these credit requests getting administered through the government schemes.

We are open for business, of course, we haven't really changed our risk appetite, but the demand, it's only going to be in the remortgage business, and those tend to be low LTV anyway, so it sort of fits our criteria quite well, but activity is much, much lower.

In terms of repayment holidays, I think there was an initial surge, and you saw that particularly in mortgages, to just take advantage of payment holidays for three months. Many folks probably didn't need them and it was something they could have done without. We've seen less of a surge in credit cards, interestingly enough. Again, I think the government schemes, things like furloughs and what have you, have calmed down consumers' anxiety. As these were announced, we saw the cancelling of direct debits or payment holidays start declining quite quickly.

I think it goes back to the earlier point. As we get out of the lockdown, we'll know what the macro world feels like, what the recovery feels like, and I think government policy will adapt to capture that. Whether there'll be other schemes available to help consumers through, or indeed the banks themselves, we may again do things that are helpful to those customers. But it's probably one for as we get into the third quarter. We won't know a whole load before then, I don't think.

Martin Leitgeib, Goldman Sachs

Could I follow up with what percentage of UK cards are currently under repayment holidays? Is it around 3% to 5%?

Tushar Morzaria, Group Finance Director

Yes, it's about 3% at the back end of April.

Okay, I think we'll wrap it up there. I know everybody's got a busy day, so I'll let you guys carry on, but appreciate you making the time. Hope that was helpful.

Important Notice

The terms Barclays or Group refer to Barclays PLC together with its subsidiaries. The information, statements and opinions contained in this presentation do not constitute a public offer under any applicable legislation, an offer to sell or solicitation of any offer to buy any securities or financial

13

instruments, or any advice or recommendation with respect to such securities or other financial instruments.

Information relating to:

• regulatory capital, leverage, liquidity and resolution is based on Barclays' interpretation of applicable rules and regulation as currently in force and implemented in the UK, including, but not limited to, CRD IV (as amended by CRD V applicable as at the reporting date) and CRR (as amended by CRR II applicable as at the reporting date) texts and any applicable delegated acts, implementing acts or technical standards. All such regulatory requirements are subject to changing
• MREL is based on Barclays' understanding of the Bank of England's policy statement on "The Bank of England's approach to setting a minimum requirement for own funds and eligible liabilities (MREL)" published in June 2018, updating the Bank of England's November 2016 policy statement, and thenon-binding indicative MREL requirements communicated to Barclays by the Bank of England. Binding future MREL requirements remain subject to change including at the conclusion of the transitional period, as determined by the Bank of England, taking into account a number of factors as described in the policy statement and as a result of the finalisation of international and European MREL/TLAC requirements;
• future regulatory capital, liquidity, funding and/or MREL, includingforward-looking illustration, are provided for illustrative purposes only and are not forecasts of Barclays' results of operations or capital position or otherwise. Illustrations regarding the capital flight path, end-state capital evoluation and exepectations and MREL build are based on certain assumptions applicable at the date of publication only which cannot be assured and are subject to change. The Bank of England will review the MREL calibration by the end of 2020, including assessing the proposal for Pillar 2A recapitalisation, which may drive a different 1 January 2022 MREL requirement than currently proposed. The Pillar 2A requirement is subject to at least annual review.

Forward-looking Statements

This document contains certain forward-looking statements within the meaning of Section 21E of the US Securities Exchange Act of 1934, as amended, and Section 27A of the US Securities Act of 1933, as amended, with respect to the Group. Barclays cautions readers that no forward-looking statement is a guarantee of future performance and that actual results or other financial condition or performance measures could differ materially from those contained in the forward-looking statements. These forward- looking statements can be identified by the fact that they do not relate only to historical or current facts. Forward-looking statements sometimes use words such as 'may', 'will', 'seek', 'continue', 'aim', 'anticipate', 'target', 'projected', 'expect', 'estimate', 'intend', 'plan', 'goal', 'believe', 'achieve' or other words of similar meaning. Forward-looking statements can be made in writing but also may be made verbally by members of the management of the Group (including, without limitation, during management presentations to financial analysts) in connection with this document. Examples of forward-looking statements include, among others, statements or guidance regarding or relating to the Group's future financial position, income growth, assets, impairment charges, provisions, business strategy, capital, leverage and other regulatory ratios, payment of dividends (including dividend payout ratios and expected payment strategies), projected levels of growth in the banking and financial markets, projected costs or savings, any commitments and targets, estimates of capital expenditures, plans and objectives for future operations, projected employee numbers, IFRS impacts and other statements that are not historical fact. By their nature, forward-looking statements involve risk and uncertainty because they relate to future events and circumstances. The forward-looking statements speak only as at the date on

14

which they are made and such statements may be affected by changes in legislation, the development of standards and interpretations under IFRS, including evolving practices with regard to the interpretation and application of accounting and regulatory standards, the outcome of current and future legal proceedings and regulatory investigations, future levels of conduct provisions, the policies and actions of governmental and regulatory authorities, geopolitical risks and the impact of competition. In addition, factors including (but not limited to) the following may have an effect: capital, leverage and other regulatory rules applicable to past, current and future periods; UK, US, Eurozone and global macroeconomic and business conditions; the effects of any volatility in credit markets; market related risks such as changes in interest rates and foreign exchange rates; effects of changes in valuation of credit market exposures; changes in valuation of issued securities; volatility in capital markets; changes in credit ratings of any entity within the Group or any securities issued by such entities; direct and indirect impacts of the coronavirus (COVID-19) pandemic; instability as a result of the exit by the UK from the European Union and the disruption that may subsequently result in the UK and globally; and the success of future acquisitions, disposals and other strategic transactions. A number of these influences and factors are beyond the Group's control. As a result, the Group's actual financial position, future results, dividend payments, capital, leverage or other regulatory ratios or other financial and non-financial metrics or performance measures may differ materially from the statements or guidance set forth in the Group's forward-looking statements. Additional risks and factors which may impact the Group's future financial condition and performance are identified in our filings with the SEC (including, without limitation, our Annual Report on Form 20-F for the fiscal year ended 31 December 2019 and our Q1 2020 Results Announcement for the three months ended 31 March 2020 filed on Form 6-K), which are available on the SEC's website at www.sec.gov.

Subject to our obligations under the applicable laws and regulations of any relevant jurisdiction, (including, without limitation, the UK and the US), in relation to disclosure and ongoing information, we undertake no obligation to update publicly or revise any forward-looking statements, whether as a result of new information, future events or otherwise.

15