Coronavirus Blew Up Your Forecast – Now What?

bit different. You may or may not have noticed that we do every episode based on a specific piece of research that BENF users can then go dig into to get the details. Today, we won't be doing that, as you might expect. The bulk of our research right now is either about COVID nineteen or its impact on the industries we cover, So we figured it would be useful to do a show about process, about how we started to look at this monstrous thing that that's voided every

forecast out there. Today, Dana and I talked with Albert Chung, head of research for BENF. He'll walk us through how we started to get our heads around the problem and frameworks BNF analysts developed to explain what's happening in to accommodate change. He also describes how all this will show up in our research, what to look out for, and

what to make of it. If you'd like to follow our ongoing COVID coverage, including our popular COVID Indicators reports, head over to benf dot com slash series, where you can subscribe to any of our fifty plus recurring publications. When a new report is published, you'll get an alert the second it comes out as a reminder beif does not provide investment of strategy advice, and you can hear a full disclaimer at the end of the show. I'm Mark Taylor. You're with Aana Perkins and you're listening to

Switch on the BENF podcast. Hi Albert, thanks for your joining us today. Hey Mark, thanks for having me. So when did it become apparent that COVID nineteen was going to become a dominant piece of the BIS research agenda. Like a lot of things, it was very gradual and then very sudden. I actually first regrete about this mystery illness in Juhan sometime in mid December, but at that

stage it didn't seem that serious. It wasn't so connecting humans humans, so really it was probably late January, when our entire team in China was asked to work from home and parts of China went into lockdown. Our team started being infected. One person was actually placed into quarantine,

so it became very real for us. And that's when our analysts in China actually really started thinking about what the impacts of the COVID nineteen was, and they put out this piece actually in mid February that said, assuming the outbreak it's contained in Q one, then GP growth will slow from six percent to about five percent this year in China, And clearly, with hindsight that now seems far, far too optimistic. And the story stayed at Asia specific

story for quite a while. Even in early March, in the global team, we were talking about impact on wind and solar and we weren't really sure if there was going to be an impact. It was the middle of March when it became really sort of unavoidably obvious that we were going to have to rework our forecasts, rework our research agenda, because this thing was going to really dominate our work in the fores civil future. So you

realized you had to change attack. Things are being quickly, perhaps quickly every seen before in our circuit at least, how did you start to get your head around situation? How did you get started? Well, I can't take any credit at all for this. Our analyst teams, particularly teams in Asia, jumped on this very quickly, and I think very rightly said Look, the first thing we need to do is to the best that we can track what's

actually happening on the ground. We launched a series of indicator reports, so we call them are COVID nineteen indicators, and they cover a range of different useful data points that we've managed to gather from different markets to keep a near real time tab on how the pandemic is developing and the impact that it's having in different markets. These indication reports go out weekly. We have one that

covers road transport. We're actually using live data from tom Tom The tracks road congestion, which is not exactly traffic. It's a measure of how slowly traffic is moving, which

is a proxy for how much traffic there is. As of the latest report that we've put out, Beijing and Shanghai are actually seeing rising congestion levels in the last few weeks now seem to be approaching roughly normal levels of congestion, whereas places like Hong Kong and Guangzo are still below normal levels and places like New Dahi and Mumbai those numbers are eighty nine down on normal levels. We've got indicators in aviation indicators and the power sector.

How do we decide which industries were going to make the cut as an indicator in which one weren't. From a sort of energy transition perspective and energy at large, clearly we're interested in demand for energy commodities first and foremost. And the first thing that happened was people stopped moving around. And when people start moving around, that's an oil story

because liquid fields get used to transport. But over time, as we've seen the lockdowns in different countries proceed with people staying at home more and businesses shutting down as well, that started to leak into the power and gas center as well. As an example here in Great Britain, we're running at about fift below normal expectations in terms of power demand on the grid. In India it's down and as much as fifty down in places Punjaba and Rochester

in particular states within India. So we're tracking really anything that's related to energy. But those indicators are having to do with the here and now, right, how do those intern impact our longer TAN forecast. Yeah, So the first revision we made was in our SOLO forecast. We've essentially deferred a chunk of the project pipeline from particularly in China, because there's been some policy changes which mean that projects

are likely to blow over into the next year. That means that we've actually revised our SOLO forecast for this year down by eight percent, and that could mean that this year is the first time solar installations actually decline like SOLO has never not grown before. This could be the first year on the wind side. Actually, we thought this year was going to be an all time record

for installations. Now we've cut that wind forecast by twelve again differring projects in We're still forecasting a record year at the moment, but quite obvious that there's downside risk to that. And just yesterday, I believe our US gas team put out a new forecast for US gas storage and they're now calling one billion q gas and storage by the end of next winter up compared to a sort of no COVID based case, which is effectively a much much loosener market than we have previously been expecting.

So those are just sort of three things on the agenda. But we're updating all of our forecasts, and I don't think any of them are going to stay the same

after the next few weeks. So you started with a delay in the most basic case, I guess, and then you move into how these things have impacked every part of our forecast, I guess, into the more structural changes, I guess you could say, so, not only does COVID nineteen really change everything for these industries and therefore all of our forecasts into the future, but to some extent, it may also change the way that we talk about everything, because from a time standpoint, it seems like things are

changing so quickly that a usual quarterly, half yearly may not be enough. So how do you decide how quickly and how often you actually need to update on some of the industries that we're doing these indicators for. Yeah, I mean, historically we've served a few different audiences with

different time horizons. In terms of the commodity markets, how a gas oil, we need to be really timely in our analysis, so we've been doing weeklies in many of those markets, for a long time, and clearly with COVID nineteen, that kind of real time analysis becomes even more important because new information comes to light that changes our view every day, and so we've got to do that really rapidly.

And then in our other sectors, if you think about renewable energy, electric vehicles, and so on, we normally update these numbers quarterly. Some of the EV sales numbers we look at monthly. But the situation changing so quickly that we just had to respond more quickly. And I think that's the same challenge that all businesses are having right now. It's the flow of news that's coming through every single day. Just got to keep monstoring it and keep on all sides.

Forecasts are tough in the best of times, and we like to do scenarios rather than forecast per se. But even that people misconstrue those is like, Okay, we're talking a bit about the future and what could happen there? What point do we have enough information to actually be

able to reasonably talk about the future. And I think, you know, maybe just even rewinding, if we're looking at this January February March timeline from the point at which we started to see changes in human behavior and changes in company's behavior due to a lot more people being

at home and government regulations. At what point were we able to start to say something meaningful in China, and then in Europe, and then maybe in the America's I think everybody has had to throw away that any forecast that was made before about February March this year, they're all wrong. And anyone who tells you that they've got a forecast that you can invest against or really believe in, it's just not really credible. And so that's a big part of the reason why we're incorporating a set of

scenarios into our analysis right now. So you knew this situation would keep evolving and you need to say smart things about it. So you introduced this thing called scenarios to be able to accommodate different things that could happen. Could you tell us a bit more about those and how those came to be? Yeah, sure, So what was extremely obvious was that any forecast made over the last

few months probably needs to be thrown away. And so when you're in that situation extreme uncertainty, which we all are in, it's better from my point of view, to embrace down certain, to acknowledge it, and actually try and incorporate into your thinking. Scenarios a great way of doing that.

So essentially what we did was we said, we're gonna come up with three scenarios for how the pandemic might proceed globally and make sure that in all of our analysis across the n F we think about all these three scenarios and how they might impact the sectors that we cover. And that's really powerful, but that acknowledges the fact that we simply don't know. Nobody knows how the pannem is going to play out. But it also means that our forecasts will have a bit more resilience to

the changing news. Every day we can monitor the news and talk about which of our scenarios looks like it's becoming more or less likely as the pandemic develops. So, in terms of how they're built, the main thing is scenarios all about what are the key unknowns that you want to explore, and so we wanted to explore really the sort of medical and epidemiological unknowns which are around how is the virus going to develop? Key things like will it fade in warm and weather? Will it come

back for a second way next autumn or winter? Will treatments and vaccines be found? What are the real infection rates? For example, are they really as low as we are being told or is it more of the population already has it, which would actually be good news in the long term. How quickly can countries ramp up their mass testing and contact tracing programs, which would then mean that we can ease some of the social distancing measures that

are currently in place. All of those things are unknown and therefore make for really good variables in a scenario, So we package those into three scenarios that eventually lead to how long does the pandemic actually last? For where are we getting that sort of information from to put in as a variable. Because the team that you lead our energy analysts, So presumably they've not spent a lot of their career looking at different impacts of viruses and things. Yeah,

that's a good question. So we're not medical experts at all. We don't have public health people on our staff, so we're operating off the same available information that other people have. We read, for example, the Imperial College paper that was widely distributed over the last couple of weeks, we read some papers about how the ninet team Spanish flu pandemic

played out globally. We've talked to one or two people who have views as well, and we've deliberately kept our scenarios pretty open minded in terms of what you would have to believe feature them to come true, because we acknowledge that we have no real epidemiological expertise, so we just framed is sort of wide range of possibilities to

make sure that we're covering enough basis. So scenario one is a single way pandemic, which basically assumes that most countries have three months of pretty heavy social distancing measures to control their outbreaks, but that after the roughly three months, then things can start to return to normal with a ramp up in mass testing, contact tracing, and so on.

But if you believe that, then you know, if everybody has a three month outbreak, then probably by the second half of the year, most of the country's have been through that, and your economic recovery can probably start before the end of the year. So that scenario one, Scenario two is in a slightly less optimistic situation, you might have multiple waves of this pandemic, and that's what happened

in nineteen eighteen. There were three waves over the course of a year, and so we've assumed it looks a little bit like that. Countries go through sort of one year a shock. It might not be exactly in two or three waves. It might be that there's sort of many on and off cycles of social distancing that are required to keep healthcare systems running. But essentially the outcome of all of that is that this massive economic shock actually lasts into the beginning and it's only the middle

where recovery begins. And this is a bit bleaker because government stimulus can keep businesses running for three months, and there's a three month shock, you can keep people in their jobs. But if it's a one year shock, that starts to look much harder, and so you're looking at more businesses going bankrupt, workers being displaced, and so on. And then scenario three years even worse. We call it

enduring pandemic scenario. And in that scenario, this thing drags on for eighteen months and the end comes when a treatment is scaled up or a vaccine is scaled up. In this scenario, nothing else works until one of those two things arrives, and that happens sometime around eighteen none from now, which pushes out the economic recovery out towards the latter half of probably the end we're really talking about the depression at that point, you know, really not

a happy place from an economic standpoint. I mean, I know this isn't a question maybe we should answer, but there's been effort. Do you have a view on which scenario could be which seems most realistic? Yeah, that's a really tough question. A couple of observations. One is, I think scenario one, which is currently our most optimistic one, was probably the most pessimistic scenario that most economists would have even entertained a month ago or two months ago.

And so I think where we are in terms of like the consensus view, if you talk to forecasters out there, I would probably guess that most people think we're in scenario one already, and the signposts are pointing towards scenario to I'll give you a couple of examples. Both Singapore and Hong Kong, which had early outbreaks and got them under control, have both increased their intervention measures in the last week or two, which to me is a signpost that says even if you control it at first, it

doesn't mean you're out of the words. It seems like through other things, like they're having important cases from other countries, it seems like they have to reintroduce measures. So that to me is a signpost towards scenario too. And how will you show up in our work? Like so, if I'm a user looking at BINGF researcher, I mean you said it's going to prevate like all of a report in the coming months. Plus should I be looking for three lines you each chart? Or how should I be

looking out for this? Wherever we have a forecast, I'm encouraging our teams to use these three scenarios. Now, the forecast revisions I mentioned earlier on solo wind and US casts, those are all scenario one forecast, single way pandemic, three month shock, end of your recovery. What you'll see as a client in the coming month or so is that more of our research will explore all three of the possibilities. Some of it will be qualitated, so it might not

be aligned on a chart. It might be a commentary on the impacts of these three possible scenarios. Some of it will be quantity too, And you'll see minds on tcharts and so on. And at the same time we'll be working closely with our colleagues in bluemog Economics. They are also updating their scenarios and we'll be working very clessly with them and try to understand how we can make the best use of their work in our b

network as well. I've already started to see some of the initial data that leads to some changes also in emissions, which is in many ways linked to the sectors that we cover as well. Is there anything you have initially started to see there and how does it change in emissions play out along these three scenarios, because there is a big difference between a blip and something that may be a bit longer term. Yeah, I mean emissions are

definitely going to be down this year essentially, pretty substantially. Clearly, that's nothing to celebrate. Stopping the economy and everybody's staying at home forever. It is not the solution to climbachation. So let's there's a school of thought that says we've now already seen the absolute peak of man made emissions, will now have been the peak of emissions, and we'll never get back to that level. If that's true, we'll have probably have brought the peak forward by six or

eight years. Like I think people were sort of thinking emissions would be later in this decade, so so that would be bring it forward a long way. I think to believe that, you'd had to believe that the shock lasts for more than a few months. I think you're talking about our second or third scenarios, because if you have a rapid economic recovery, I think emissions would be back to previous trajectory. I think you'd be back onto

this growth curve. If, on the other hand, we're looking at a more prolonged pandemic and longer recession, I think we might find that the low carbon transition might have proceeded further before the recovery really starts to take hold. So, for example, if a load of high carbon infrastructure gets retired or mothballs during a recession or a depression, and in the meantime the costs of clean alternatives continue to decline, then you might not see some of that legacy high

carbon infrastructure come back online when the recovery begins. The shot onneswers nobody knows, but it's a really valid question. I think we're going to think more about that. Well, COVID nineteen is nothing to celebrate if we look at emissions in isolation. The fact that we might have reached our peak is something that could potentially be a very good thing. As we're headed into a period where there's going to need to be stimulus in economies around the world.

There are rumors flying about that some of this could be tied to different green incentives. What have you heard about green stimulus to date? Surprisingly, there's been a lot of talk about green students. And I'd say that surprising because the focus right now is rightly on trying to keep businesses from going bankrupt, try to keep people employed, and keep the house and fair, and deal with the

healthcare situation. And if this turns out to be a deep but short recession and recovery will be a lot easier business stay afloat. And that's the right focus, and that's what we're seeing here in the UK, for example, there's this three billion pound Learned Guarantee program and the government's committing to cover employees wages if they're furloughed, etcetera. And the US has two trillion dollar stimulus package, and all of that is rightly focused on the short term.

Having said that, clearly there's a lot of people now talking about what recovery packages could look like when the time comes to start stimulating the economy again, and how much of that should be green. So I think there are some opportunities to push a sort of climate agenda in any recovery package. So the things that lend themselves really well to stimulus are projects that are quote unquote shovel ready, they're ready to go, ready to ramp up,

and projects that create jobs. So I think things like renewable power, energy storage, EV charging infrastructure, those are supply chains that are ready to ramp up if they have project developers that are ready to move forward with sites in some cases that are already identified. So in terms of job creation, there will be quick wins, and there'll be quick wins for the climate as well, energy efficiency as well as a few one we could make huge

improvements to our building stock. I would bake in huge carbon reductions for the coming decade, and that would create a lot of jobs in local communities as well. And I'd also love to see some capital going into circular economy for example, I mean, I could see increase investment

into recycling capacity. So I think there are a lot of things that people can see aren't doable, and I think that the early conversations we're seeing now are about positioning some of those conversations so that they're ready to go when we're ready to really talk about recovery. So a lot of our clients use our research as a guide or a data point or an input, but then they have to still go and do their own to figure out how things are going to impact their company

and their strategy going forward. Is there anything that you found in doing your researchers setting the agenda for BNS research that you'd recommend as they try to get their head around the impact of COVID nineteen and the impact in their company. The companies I've spoken to, the clients I've spoken to are already asking most of the right questions and thinking about the right issues first and foremost worrying about that people how to keep and safe. They're

thinking about keeping their operations running wherever possible. They're thinking crucially about their suppliers, their supplier supplies and the whole supply chain, and whether those companies need help, because that's quite hard because the visibility isn't always there. I've spoken to auto companies, for example, who are really working to try and figure out further upstream which companies and countries are actually exposed to that they hadn't necessarily had to

think about before. But I think companies are figuring that out. I think the only thing I would add is to really think about signposts as a way of navigating uncertainty. We've developed our three scenarios, so the key thing for us now is every day every week to read the news, look at what's happening, and ask ourselves, does today's news move us more towards scenario one or two or three? And how should that change what we do in our business.

I think every company will have slightly different scenarios and slightly different signposts they should be looking for. But my advice would be too carefully and deliberately define those and pay attention to them over the coming months. So it's lair. The future is uncertain, but we're tracking a number of things to make sure that we're staying on top of how this story of all of our lives is constantly evolving.

So Albert, thank you very much for giving us a little bit of insight today into the research process and how we really have had to change how we look at the future because of everything that's happening right now. Albert, thank you for joining Mark and I today pleasure. Thank you for having me. Bloomberg an e F is a

service provided by Bloomberg Finance LP and its affiliates. This recording does not constitute, nor should it be construed as investment advice, investment recommendations, or a recommendation as to an investment or other strategy. Bloombergun e F should not be considered as information sufficient upon which to base an investment decision.

Neither Bloomberg Finance LP nor any of its affiliates makes any representation or warranty as to the accuracy or completeness of the information contained in this recording, and any liability as a result of this recording is expressly disclaimed.