 Welcome to a coronavirus briefing with MIT. I'm Randall Wright, program director with MIT's Office of Corporate Relations. I'll act as your host on behalf of MIT's industry liaison program. The coronavirus outbreak is first and foremost a human tragedy affecting hundreds of thousands of people. It's also having a growing impact on the global economy. This briefing is intended to provide business leaders with a perspective on the evolving situation, implications for their companies, and steps they should be taking now to mitigate the impact of the COVID-19 pandemic. We have with us today three of MIT's most prominent faculty members, Professors Joseph Sheffey, Alex Pentland, and Andrew Lowe, to give us perspectives on supply chain, organization structure, and financial markets, and insight on what firms should be doing now. Let me begin by introducing Dr. Joseph Sheffey, Professor of Engineering at MIT, and director of the MIT Center for Transportation and Logistics. Today, he will focus on the here and now, and will explore with us the COVID-19 disruption and context of past disruptions, and explain what companies should be doing now as the epidemic is spreading. Professor Sheffey. Hello, everybody. I'm Joseph Sheffey. I'm the director of the MIT Center for Transportation Logistics. In the next hour, we'll talk about business and supply chain impacts of the COVID-19. Of course, the first thing that people tell you is not to panic. And I'm not panicking because my computer is not working. So just a minute. There's a technical difficulty here. Yeah, just click on the slides once with your mouse. OK, we got it back. So the first thing everybody tell you is not to panic, which, of course, is the reason for all the panic. But let's look at the whole disruption that business are facing here in the context of business disruptions in general. So we have random phenomena like hurricanes and floods and natural phenomena. We have accidents. We have government and politics, Brexit and lots of other trade disputes and others. There are issues of non-compliance. We probably don't comply with regulations. And it brings a disruption of business. Or a supplier does not comply with regulation. We have sometimes competition coming from left field without us realizing it. Of course, the economy. We're now going towards a recession. But we've had recessions before. 2008 comes to mind. It was a big disruption for all businesses. There's issues of social discontent when people are angry about using, for example, animals for testing, about the environmental issue. There are, of course, intentional disruptions, be it strikes, be it terrorist attack. And finally, we're going to talk about pandemics. That's the subject of today. So in order to put it in context, let's look at some history of some pandemics. So the AIDS pandemic, which is still slowly going on, had 75 million cases at a death rate of over 40%. The US flew just in the US in the last year at 35 million people infected with about 0.1% fatality rate. The H1N1 was a very big disruption. Had about a billion people worldwide. About 200,000 died. The Ebola is one of the more frightening one because of 2-thirds of the people infected died. Finally, we're talking about the Spanish flu, which several people mentioned in 1918. About 1.2 billion people were infected. 60 million people saying 50 million, 70 million. Nobody really know for sure. About 5% fatality rate. But the numbers were enormous. And this was what hospitals look like. And this is where mass graves were digs. So let's now talk about how pandemics work. And they don't attack the whole world at once. There's a first cluster going on, what we saw in China. It then dies down. Then we have another cluster starting and dies down and another cluster starting. That's what it looks like today. And if we look at one of the dashboards, you see the clusters happening every day. Also, if you look at the lower right-hand side of the screen, you see the top line, the orange line, is infection in China. The yellow line that goes like a hockey stick are infection in the rest of the world. Italy and, of course, around Europe in general. And right now in the United States. Just to understand, there's a lot of misinformation about the fatality ratio. So let's me just mention it. The problem is there are two ways to calculate the fatality ratio. The first one is for countries who are caught flat-footed. They didn't test in time, many undiagnosed infection, and the results are overwhelmed medical facilities. So this is what happened in Wuhan in China. This is what is happening right now in Italy. This is what is happening right now in France. In China, just to understand, during the Edehubai province, there were about 4.5% people died. However, in the rest of China, it was less than 1%. Because by then, they were implementing all these measures to stall the pandemic. Around the world, you can see the difference between countries. In Italy, it's now over 7%. The number of dead people in Italy is staggering. South Korea is less than 1% because they started very quickly testing people, following those that came in touch with those that became positive. And then they have a very low mortality rate. Iran, almost 5%, has a problem. But if you look at Switzerland or Denmark, they have very low percentage. Because they took a lot of the measures early. The idea is to take it early. And just by comparison, there are several countries that took very aggressive measures early and experienced no death. Whether it's Malaysia, Qatar, Singapore, Israel, Finland, Bahrain, none of this country. There are hundreds of infections in this country. None of them experienced any death yet. Let's now move to the subject of this talk, supply chain impact. So first of all, comparisons between this disruption and others like the Fukushima meltdown, the Thailand floods, the SARS, MERS, Katrina, and the United States are all inaccurate and underestimating the impact. The problem with the COVID-191, COVID-19, are that it impacts both supply and demand. We saw first a supply impact. And I'll talk about it a little more. But the main impact is going to be a demand impact. Part of course, as you read in the paper, China today is much larger element of the world GDP. So both supply and demand affecting China, thereby affecting the rest of the world. But the main impact will be the demand reduction due to fear and social distancing. And of course, the media, the online hype. We are probably facing a recession. I let the two other speakers talk about it a lot more. But this is the so-called Black Swan, the thing that hit us from left field without any preparation. Let me suggest that even though this is different, there's something that I call the Anakarenina principle. What is this? Tolesta wrote in his book, Anakarenina, happy families are all alike. Every unhappy family is unhappy in its own way. So every disruption comes with its own litany of misery, causes and cascade of effect. Every disruption, not two disruptions are the same. However, management of every risk and every disruption involve prevention, detection, and response. There are generic preparation steps that should be taken in every disruption. These are applicable to supply chain issues and we will talk about it right now. So first of all, the first thing that I want to explain is the so-called bullwhip effect. It tells us who will be the most vulnerable among businesses affected by the supply chain. What happens with the bullwhip effect is as follows. Assume that a retailer sees consumer demand go down by, call it 10%. The retailers do think now, it thinks that the demand will be down by 10%. I'm just using an example number. We'll go by 10% for a long while. It also has inventory, assuming that it will be 100%. So it has to reduce for the future, its order, and it has to get its inventory down. So instead of just ordering from the distributor, 10% less, its order may be 15 or 20% less. The distributors see 15 or 20%, it has the same issue. It looks at its forecast, it looks at its inventory, and its order from the producer may be 25% less. The producer's order from supplier even less and it goes up and up. The bullwhip goes up and down. This point we're going down. It also, you know, also laid heavily. But the point is that the small suppliers, the second and third tier of the supply chain are usually small and cannot withstand the initial downdraft in orders. So the result is damage to the upstream suppliers. Just to give you some data, during the height of the 2008 crisis, which was also a demand disruption. US retail sales were down 12%, manufacturing inventory down 15%, and manufacturing sales were down 30%. A colleague of ours, Jan Fransow in Holland, did a study with using Dutch data during 2008. Tier one or two is the suppliers immediately supplying to the brand owners. Relative to the end consumer, they went down about 25%. However, people deeper in the supply chain, what we say, what we call upstream, relative to the end customers were down 40%. And in fact, the Chinese immediately understood as the China Morning Post, that this could be a death blow for Chinese small manufacturers. So the Chinese government told state bank to start loaning money at zero to very low rates, sometimes zero to small businesses and reduce aggressively the taxes on all small manufacturing. Now, during 2008, Ellen Mulally, the CEO of Ford, understood it very well. When he gave his testimony to Congress, he did something that in some sense is amazing. He implored Congress to save its competitors because he said if GM or Chrysler will fail, their supplier will fail, their supplier supplier will fail and their suppliers will fail. These suppliers are making, for example, Johnson Control makes seats, makes seats for all the auto manufacturers. If it fails, nobody will have seats, including Ford. So it asked Congress to help and make sure that these companies will be held. Ford was not in a bad position, it had enough money. But to summarize, you say a collapse of one of our competitors would have a ripple effect across all automakers, suppliers and dealer, a loss of nearly three million job in the first year. So he was trying, he understood this problem very well and trying to explain to Congress that it's not only the, what you call the OEM, the original equipment manufacturer, the people who hold the brand. In this case, it's the other company, but it's all their supplier and it will affect an entire industry. We'll talk a little more about this later. First of all, let me give you some forecasts, which is always dangerous because in two to three weeks, you may find out that I was wrong, but be it as it may. So we started in, let's talk about supplies from China. First of all, I understand that supplies are coming by and large the big quantities on ships. They are not flying on air in any case. There's no, almost no flying, but this is not flying on air. It is coming on ships. So what happened is the following. In the Chinese New Year, there's all businesses in China almost grinds to a halt. It goes from the second week of January until the end of January, beginning of February. So this is what happened in 2020. Everybody went home. Factories, the world over who gets part and material from China knew not to expect shipment during the Chinese factories closure. So, however, they expected shipment to restart right afterwards. So what we did, what these world factories did all over the world, they had enough inventory of parts and material in order to build their product to cover the down period, but not much more. So see what happened now. Let's focus on China and its vicinity. What happened is a week after the end of the end of the Chinese New Year, ships from China loaded with parts and material did not make it to North Korea to South Korea. They probably also didn't make it to North Korea but it was another problem. It did not make it to South Korea. What happened in Hyundai announced a week after the end of the Chinese New Year that is closing factories in South Korea. 10 days after the end of the Chinese New Year, ships laden with parts and material were not making it to Japan. Nissan announced closure of factories in Japan because it did not have parts. Let's now look at the United States. It takes about six, eight weeks to get from a factory of China to a factory in the United States. What does this mean? This means that if we take about the beginning of February, sometimes the end of March, the first part of April, we're gonna start having factory closures in the United States. Whether it's automobile parts, whether it's cleaning supplies, whether it's aircraft parts, whatever. Many, many others. We're gonna start having closures and inability to make product in the United States for a while until the Chinese will start moving again. But remember, it will be not possible to fly many of these parts because we lost a lot of the air cargo capacity. Just for those who not realize it, most of the air cargo capacity is belly freight, what's called belly cargo that flies in the belly of passenger aircraft. Most passenger aircraft are not flying to China. In addition, FedEx, UPS, DIGL all reduce their China services. So we don't even have capacity to do it, even if we wanted it, it would be outrageously expensive anyway. So it will be a while until these factories will start gearing up. So what are corporate, what should be the corporate actions that the companies should take right now? And we'll talk about setting emergency management centers. We'll talk about how to set up communication decision-making protocol. We'll talk about who has the decision-making authority, reviewing the review of suppliers and review of products and customers. And finally, well, one bullet point before last, we'll talk about one issue about finance that is tied to supply chain, note that the Andrew Law, my colleagues, will have a whole hour talking about financial issues in the context of the coronavirus. And the last one, we should plan for the recovery. We should plan for being able to get up and go at the end of this and we'll just mention what should be done then. So first, let's talk about emergency operation center. So the idea is, when you have a crisis, you want central information and decision-making function. You want the decision makers have all the information come to one place when people can talk to each other and make decisions. Now, the pictures that you see here are, or the illustration that you see here are probably not relevant for a coronavirus because a lot of this, in case of the coronavirus, will be done virtually. But we need a central information decision-making and one can have people in a room like this, but spread them around, of course. In this type of emergency operation center, you have to worry about two things. You have to worry about employees. How do you take care of employees? And how do you take care of the business ecosystem? And in many ways, this should be two separate teams because the mistake that many companies are doing is one team looks at both and you have to worry about both of them. Both of them are as important at the same time. When you talk about employees, you have to think about families, you have to think about children at home, you have to think about how to connect employees, our continuity of pay, a lot of these issues have to be dealt by the HR team, basically tied to the emergency operation center. Of course, taking care of the business ecosystem is what we'll talk about later. How do you take care of customers? How do you take care of employees? How do you take care of suppliers and so forth? And lastly, taking care of communities. Corporations operate in certain communities and have to make sure that they offer their services to the community. So for example, Google made its online tools for collaboration free for businesses for the next three months or so. Many businesses are taking similar actions, giving their staff free to people so they can keep business, family, relationship, collaboration going during this crisis. Just to give you an example, this is the Walmart emergency operation center during a hurricane. You see people sitting next to each other. That's not gonna happen now. They're gonna sit far away from each other, but this is tied to all their system. On the screen, you can see what happens in every one of Walmart distribution center and stores and parking lot. And the information comes to this place. This is American Airline operation center, something similar. Let's now talk about the crisis communication. Excuse me. How do you communicate what has to be done in case of a crisis? So first of all, you must have ready communication for all stakeholders, which means know who they are, have all the emails, have everybody get on similar collaboration tools at this point and so on. As I mentioned before, you have to worry about employees. You have to worry about customers. You have to worry suppliers, the media, shareholders, analysts, and the community. You have to communicate with them continuously. For example, MIT every morning sends to all its employee a message about what's going on and how life are changing, what is happening, what is not happening. This is something and it's sent. Now of course, MIT, everybody at MIT has an email address and tags and so forth and it comes on all channels. Important to speak with one voice. Just one person speaks about the crisis and what we are doing about the crisis. Of course, we have to decide who is in charge. We all remember after President Reagan was shot that Alexander Haig came on TV and saying, I'm in charge. Turns out according to the constitution, he is not. But so one had to decide a priori what happens if the CEO falls ill? What happens if other people in the organization are not able to function? Give accurate information. This is something that we all lament now in the United States, but giving accurate information. Tell exactly what's going on. The one voice has to tell everybody exactly what's going on not try to spin it, not trying to blame other people. Just tell the truth. Example of how not to communicate is the case with Malaysia airline, the Boeing 777 that went into the Indian Ocean. On March 8, 2014, Malaysian airline 370 flight disappeared. Immediately, the Malaysian went into a tizzy. They started conflicting announcements and spoke with multiple voices. See, the Malaysian Prime Minister was saying one thing, the Minister of Transport and other, the Inspector General of the Malaysian Police, the Malaysian Air Force, the Maritime Enforcement Energy Department of Civil Aviation, airline representative, air traffic control releases, all gave very confusing information and nobody had the trust of the people and of course the grieving relatives. But also the whole aviation work tried to understand all aviation world tried to understand what was going on. What happened to a sophisticated state of the art airplane that fell off the sky? We still don't know. And of course we have a problem. When we talk about how not to communicate in the US with the president and I'm wondering what the people around him are thinking. I think that this guy doesn't really really, doesn't really believe what the president is saying and this guy gets really angry. So I don't really know what they're thinking. That's in my mind what they're thinking. Now, decision-making. Let me give you an example that I described in my latest book about risk management and resilience. In 2011, that was the height of the Japan disaster. Remember the tsunami, the earthquake, the tsunami and the nuclear disaster. So many of the GM suppliers in Japan were not able to supply parts. So for example, at one point, GM had no seat heating module. That's the solenoid that goes under the seat to warm it up in a cold weather. So they had no seating, couldn't get the part that hits the seat. So they had an emergency management center, of course, but they were arguing what to do. A vice president came in and made a decision, build vehicle without heated seats. Well, turns out a disaster. The problem was that heated seats go with leather. So you build more cloth seats. So, but the cloth and leather mix affect the basic versus luxury model. So what happened is canceling the leather seat meant all sub-assembly and component that went into the seats became stranded in the supply chain somewhere. And anybody who works in supply chain know that the worst thing is to have too much stuff going to a distribution center or a warehouse. And you had all the leather seat clogging the supply chain. They had it with a few other parts. On top of it, you said you're gonna build more basic and basic cars with cloth seats, but dealers and customers, what they want? They don't care about GM problems and they're just not gonna buy. So the idea is, now, of course, if you are an engineer or a supply chain professional, you know these things and you will never make a decision like this. So the mantra, one of the main conclusions from this crisis GM that then they implemented everywhere is called swim your lane. Which means, in other words, that's what Bill Belichick, our beloved coach, calls do your job. Don't do other people's job. Don't swim in other people's lane. Just do your job. If you are an engineer who deals with seats, you know what can and cannot be done. If you are supply chain manager who understand where stuff is coming from and where it's going, you make the decision. One of the most important issues in this case is actually to keep the C-suite, the CEO and others away from making decisions. They should talk to the media. They should talk to analysts and they should be fully informed all the time about the latest information, but they should not make the detailed decision because you talk about sophisticated product like automobile with tens of thousands of parts and tens of thousands to hundreds of thousands of suppliers all over the world. You have to know exactly what you're doing when you make decisions. So let's now talk about the next point which is look at suppliers. You want to make sure that you still have supplies, that you have a steady flow of parts and material into your factory so you can build your product. So the first thing you have to do is supplier mapping. What do I mean by supplier mapping? You have to know where your stuff is coming from. Surprisingly, most companies don't. What they know is who the supplier is and they know where the headquarters of the supplier is because in the SAP system or other enterprise research planning, what you have is the address of where to send the check where the suppliers send you stuff. That's not what you need. You need to know where the supplier plants are. Are they in Wuhan? Are they in South Korea? Are they in Northern Italy? Where are the plant of the supply? Not the fact that its headquarters is in Luxembourg or Switzerland. This does not help you. You want to know where the plants are so you know if they'll be affected. Next thing you have to know about each one of these plants of the suppliers, what do they make? What parts do they make that goes into your product and how critical these parts are? What I mean how critical, there's a lot of things that go into this but can you build your product without this part? Can you, do you have two other suppliers of this part in other parts of the world who can make it? Also very important is which product do these parts go into and which customers need this product? So now you have to think about the next point which customers are served. That's very important. You want to decide, let's say you don't have enough you have to decide which customers should be served. And before we talk about inventory level, we'll talk later about what to do when you don't have enough supply and how to decide about which customer to serve. But you have to check inventory level not only in your warehouse or plant but at your suppliers and their suppliers who has inventory, who can keep building parts and sending them to you. Capacity, are they working at full capacity? Are they working at 80% of the capacity or 50% of capacity? This will give you an idea of how many parts and how much material you can expect to come into your manufacturing facilities. And you want to have what we call visibility. This is something, this is not something new. Most companies are trying desperately to get visibility into their upstream supply chain. What's coming into them? I mean, you know what you're sending out to your customers. What is much harder is to know what's coming into you. Surprisingly enough, even though every consumer when it gets a FedEx, UPS, USPS package can follow the package and track it, most businesses cannot. Most businesses cannot follow something from a factory in China all the way on the truck, on the rail, on the ship, goes to US port, go to customs, goes on the rail, goes on a truck and get to their plant. It's very hard to follow this as it turns out. We can talk about why it's, lots of people are investing in this a lot of money. It's getting better. The problem is what happened to the secondary supplier? What happened to the stuff that has to come into your supply? Can they know what's going on? Can you know what's going on even though it's not coming to you? That's what we talk about when we mention visibility and with there are some, there's a lot of work on this with intent of things, with new sensors and new capabilities. What to watch for? Okay, lots of things to watch for in case of a disruption. Supplier may degrade quality. They may be under pressure. They want to serve you. They need the money and the stuff that they build does not have the same quality to use material that's not approved. I have to watch for you. Late delivery. They promise to send on a certain date. They just can't. So deliveries will be late. That's why you want all this visibility. A huge problem is fakes. When there's not enough supplies, when supplies dry up, companies start looking everywhere for alternative suppliers. For alternate suppliers. Lots of new suppliers suddenly come online and on the internet and they offer fake stuff. They may offer used chips, for example, that may be repainted and sold as new chips but they'll fail as soon after you install them. Lots of issues with this and we saw this big time. We saw this doing the horse meat scandal in Europe. We saw it during the US Air Force, the Lockheed and the people who built some of the most advanced just in the United States found out that they have fake chips in some of the systems. There are many, many cases like this but what's important is to watch for it during a disruption because this is when they spike. Financial health, you have to watch for the financial health of your suppliers. If you depend on them, they better be in business. So watch for, and it's not enough to wait for the end of the quarter. You want, the best thing is actually to watch the news in their city, in their area to find out if there's too much talk about redundancy and bankruptcy and going out of business. Things like this, of course, call them all the time. Be in touch with them. Finally, get ready to support critical supplier. In 2008, we had many companies, automotive companies, Intel, many others supporting weak critical suppliers by other investing in them, extending their own terms to this, their own credit terms to these suppliers and many other ways. But the important thing is make sure that critical suppliers don't go out of business because then you go out of business. You cannot build stuff. Think about it. Now, there are several problems with the supply chain structure. This is what the supply chain structure look like. On the top, you see OEM, Original Equipment No Factory. Let's assume that this is Ford Chrysler and GM. The first problem is opaqueness. Companies actually don't know who is the bottom tier, the top tier of the supply chain. If you look at the OEM, Imagine General Motors, it's no who it's tier one supplier. Let's assume that one of them is, we say Johnson Control, makes seats. The seats go into GM plant that assembles the cars. But into Johnson Control, there are companies who make all kind of parts for the seats. So this will be the heating modules that we talked about before. And upstream in the supply chain, there are lots of suppliers who make screws. And this is by the way, when you go into automotive, there are special screws that are actually regulated and heat treated, but they sell to all the auto companies. And the OEM doesn't know who is selling to them, where their stuff is coming from. You know, if the OEM know who it's tier one suppliers are, some of them would know tier two, but many tier one in many industry would not reveal who their tier two are. And it gets to be very opaque when you go upstream in the supply chain. So the first problem is it's very hard to know even what to watch for. The second problem is that there's could be vulnerability of an entire industry. For example, it could be that unbeknown to all these three tops, original equipment manufacturer, they all depend on, for some parts, on one supplier, a tier three, they don't even know who they are. We call it a diamond structure in the supply chain. Examples like this was in 2012, a factory, Avonic Chemicals, it's a factory in Germany blew up. It supply over 50% of the automotive needs, of the world needs for automotive what's called plastic 12. It's a certain type of plastic that is very resistant to corrosion, to fluids, to a lot of other stuff. It is 40 to 50 pounds of this is used in every car around the world. It was a disaster that almost brought the entire automobile industry to its knees. Turns out the industry to its credit got together, all the competitors got together under the guise of the automotive industry group. And in Detroit brought together a lot of other chemical manufacturers, DuPont and many others who started making these stuff so the auto industry did not come to a halt. I mentioned before what do we do if there's not enough supply. So if you have limited supply, you can do several things, you can allocate the goods, you can auction them, you can dilute them, and you can shape the demand. Let's talk about every one of those. Let's talk about allocation. Excuse me, let me, first of all, when we talk about allocation is product triage. In many cases, the same parts may go to many different products that you make. So again, which product gets the part? How do you do it? Based on financial contribution, which means the product with the highest margin based on product for which you have a lot of products already in the field, based on products that go to the most important customers, based on fairness to everybody, based on which customer really totally depend on you and go out of business without you. There are many, many considerations about which product gets the part. And you have to decide this now before, so you don't have to make decisions on the crisis. Of course, it works best if this consideration, like margin, vary across product and customer. If they're all the same around all products, you just do it by fairness. You give everybody the same percentage of the product, but it's never the case. Usually, it's very different between customers and product. So you give you an example again with General Motors with 2011 during the Japan crisis. So there was a shortage of engine controllers, airflow sensors, and brake control modules for all trucks. These parts are the same for all trucks that GM makes. GM made a decision. It made a decision to close the Shreveport truck plane that makes the Colorado truck. This is the truck shown here. Why do they make it? Because this is a small truck. GM makes a lot of money on big truck and very little loses some money on the small trucks. So in addition to this, it had lots of field inventory because it was not selling well. It has a lot of field inventory. So it would not have affected customers. So GM decided to close the Shreveport plane for one week. Of course, the Wall Street Journal and New York Times, so the sky's falling GM is closing plan. It was totally and utterly planned and orderly closing. As it turns out, once they decide to close it and start the process, they found out enough extra parts and did not need to close. But by the time that you start closing of a plant, they had to close it for a week and then ramp it back up. But it's the example of product reaction. An example of auction is in Thailand, France in 2011 was a huge floods in Thailand. You see all the factories being flooded. Thailand is the number one producer, like 70, 75% of the world's disk drive are produced in this part of Thailand. So companies, the company that had the part decided to get it to the customer who wanted most. Now, economists will tell you that this is the right way to make a decision. Western digital was hit hard and SIGET had demand stripping supply. So SIGET conducted auction among customers. He who pays me the most will get it. Economist thing is the right way to do it because it gives the product to the customer who needs it most. Unfortunately, the response of the customer was that this is profiteering and once they had to go to SIGET, but when the crisis abated, they all went back to Western digital. So it worked for a short term, but not for a long term. Delusion, that's an example. There are many other examples, but the maker mark had high demand for his bourbon. So he decided to add what they said, a touch more water to the burden. Instead of 90 proof, it became 84% proof. The CEO explained that to maintain the test, they had to dilute it a little bit. The result was a customer revolt. One customer wrote, my bourbon is being diluted, so beam can fatten the wallets a little more. I'll have lowered their demand by not buying anymore. Customers stop buying it. The CEO, a week later, the CEO said, we've been tremendously humbled over the last week or so. By the way, this was not the, many other company diluted many products. For example, Intel diluted some of the material that goes to the chip, but of course it was well tested and to work. It was not a problem. There's another, the last thing is substitution. During the Taiwan earthquake, both Apple and Dell introduced no model. Apple, in order to estimate demand, took orders for 160,000 specific computers, the specific component. Dell had no pre-orders. So Apple could not sell what they promised. Dell used its so-called postponement strategy. It would not build a computer until it had an order in hand. So it just raised the price of those computers that had the component in short supply, lowered the price on those, those computer for which it had all the parts and continue selling with no problem. Let me just say a word about finance and a lot more of it will be said by Andrew. First of all, during recessions, as we seem to be going into cash is king. So in terms of supply chain, you have to think about paying later to suppliers who keep cash. You have to try to get earlier payment from customer. Once you send them the product, you have to reduce inventory. There's a lot of cash tied in inventory. You have to delay any capital investments. And of course, when you increase the time you pay suppliers, watch for the health of supplier. You can do it with some suppliers, not with the weak one. Of course, make sure you secure credit line and start thinking right now about the recovery. I'm not talking more about it because Andrew Lowe will talk more about finance, but let's talk about planning for recovery. So the idea is to keep expertise in house. So continue to pay, taking care of family, allow for part-time work. Germany, Holland changed the label for example. They want to make sure that workers with expertise are not going to leave the market or go and work elsewhere. So they allow for part-time work, but the big change was that unemployment would be paid for part-time work on a part-time basis. And then let's not forget, a crisis is a terrible thing to waste. And it's opportunity during a crisis for tough business decisions. It's an opportunity for reorganization that you always wanted to do, but there was a lot of resistance within the organization. There's an opportunity to cut not performing products and customers under these circumstances. Finally, a note about the long-term because I get a lot of questions about it. Will more companies move manufacturing procurement onshore out of China, out of elsewhere? Short answer is yes, but not much. The reason is that Chinese manufacturing become so sophisticated that it's not easy to replace. And Southeast Asia still has cost advantage being Vietnam, Bangladesh, Indonesia, Malaysia. So it's not easy to move onshore. But on the margin, companies will not move totally, but will balance their supply base and include supplies in different parts of the world and maybe in the United States as well or in North America. But note, companies who source in the United States will do the same thing. They will be afraid that something happens in the United States. By the way, like it's happening right now and have some foreign dual sources, have other sources elsewhere. So company will pay to have more than one source. Some of them will be in North America. Some of them will be elsewhere. So finally, let me stop here and I'll try to take some questions if I get some. And just to just further reading, I wrote two books on supply chain disruption. After 9-11, I wrote the resilient enterprise that came out in 2005. It's called the, I don't need my glasses, but it's overcoming vulnerability for competitive advantage. And in 2015, 10 years later, many of the companies that I talked to came to me and said, look, there are a lot more risks, but they were also much better at it. We know what we're doing. So we should write another book. So I did. And in 2015, I wrote The Power of Resilience, how the best companies manage the unexpected. None of this talks about coronavirus, but a lot of analysis of past pandemics. Finally, what you see down there, HTTPSchefie.mit.edu is my website. And if you go there and look on the LinkedIn influencer post and many others, you'll see on the front page, a lot of posts, TV appearances, works with journal articles and others, stuff that I wrote about the coronavirus in the last few months. So with this, let me see if we can take some questions. Okay. There's a question from Solomon. He said, we've been pushing decentralization and pushing decision making to the lowest possible for quite some times. Are we scraping all this with the move to centralization in a crisis? Good question. Many companies have been decentralized and moving decision making down. And by the way, it's absolutely the right thing to do, especially in a crisis, but let's separate between two things. Let's separate between the information flow. We have to focus the information and focus high level decision making. But the idea, if something happened at the low level, this particular important, not for this type of disruption. In this case, it's worldwide. So knowing what's happening in other part of the world, learning from what's happening in other part of the world, concentrating this information in one place is a good thing. But then moving this information down to the decision making maker at the lower level of the organization. So in general, absolutely. Yes. Moving decision making to lower level organization is the right thing to do. But in this case, since it's affecting the entire world at once, not a particular, a particular location was to act fast. We have to learn from everything, understand what's going on around the world, move the information down. And unless it's a decision to affect the entire corporation, which has to be done centrally, decision can be made locally. So with this, let me move to the other question. Philip asks, given the information overkill at present, my God, I'm hope I'm not part of it. What are the trustworthy source of information on seaborn and airborne logistics from China to the rest of the world? Okay. Right now we have a, the idea is call your favorite supplier. Call your favorite shipping company airline that you ship it, mostly forwarders who are forwarders for those who are not, not familiar with the intermediaries that you arrange for you for either sea or air international shipping. They will have much better information because they talk to the actual airlines and shipping line all the time. So to me, the best source of information are these intermediary, the third party here in between the customer, the retailer, manufacturer, distributor who owns the goods and the customer where you want to send the goods to. In the middle there's these forwarders, these third parties who can have the information all around. So my guess is that this will be the source of this information. Okay. Unfortunately, I'm going to sign off now. I'm going to get off this now because I want to finish five minutes before the hour because we have to clean up the surfaces and I'm sitting in a very small closed, you know, audio booth and we want to clean it up and we want to air it out before the next presenter come. So let me stop here and invite Sandy to come in and follow me. So thank you very much. Go to my site. If you have a question, you can send it to my email chefie.mit.edu. My site has more information chefie.mit.edu. Thank you very much for listening.