The panelists here and their topic are focused on financial risk management, but the broader topic is really about resiliency, being prepared as a business for the expected or even the unexpected event or risk.
As Nicholas Taleb,the author of The Black Swan, points out, you never really know, especially in the real world, what is going to happen. Anything can happen, but you need to be prepared. You have to have an action plan. Taleb was an options trader; that means have a stop-loss in place. Have a threshold or a point where you protect yourself, or you get out. Risk in the treasurer’s world is really anything that affects liquidity or access to liquidity such as financial or commodity price risk and/or account or party risk among others.
Prudent companies look at risk systematically. They look at it within a complete risk management framework of economics, controls and accounting. They understand their stakeholders’ risk appetite, and they approach risk management consistently throughout the company in an enterprise approach. They have established objectives or thresholds within their risk appetite to manage their risk and prepare for the unpredictable event. They have an action plan. They have an objective.
They also understand the buy-in in risk management. The foremost buy-ins are predictability and liquidity, and the avoidance of financial distress. Prudent companies have a process for analyzing their choices in dealing with risk, based on their own risk appetite.
As you’ll see in Microsoft’s presentation, you can’t always pass the risk either back to your suppliers or on to your customers. Whether it’s based on competitive pressures in your marketplace or other issues, the risk itself doesn’t go away. It has to be managed, and the point is, who is better equipped to manage it?
Finally, prudent companies and thus resilient treasuries have established controls in place to assure that risk management objectives are carried out, and that they are carried out without causing other undue risks. Even with some of the most sophisticated technology and modeling techniques, Wall Street (and I use that term very generally) in the end only grew overconfident that they had their risks conquered. And I know I have to speak softly here because there is a risk conference going on down the hall. But their fallacy was really over-belief in their models, which turned out to be either based on flawed assumptions or, more than likely, management override.
Models, as we know, aren’t the answer. Models are tools. Models can’t really be given all the credit, and they can’t be given all the blame. To quote Taleb again, models are useful tools to provide insight. They’re an aid for critical thinking. You need to understand them to know their limitations, as a bad model is worse than no model at all. Models do provide a level of objectivity in your decision-making, however, and they do show you alternative paths--what could happen even at your maximum extremes. So they are very useful.
But even the best models carry the risk of management override. Consider again what has happened to several of the Wall Street firms. Even as the pressure got to the conservative ones, the question was, “Everyone’s making money, so why aren’t you?” The answer of model risk was no longer an acceptable reply.
Unfortunately, business leaders blatantly ignored all the warnings. When asked about the risks of the system, the former head of Citibank, Chuck Prince, infamously said, “When the music stops in terms of liquidity, things will get complicated. But as long as the music is playing you’ve got to get up and dance. And we’re dancing.”
This suggests that he knew the risks, but he chose to ignore them. You can have the best system in place, but you cannot always thwart inefficient human actions and behavior. The financial crisis has given us renewed visibility to the treasurer’s role and the importance of protecting liquidity. It has brought a renewed focus on risk management and the importance of being prepared.
Financial Risk Management 2010 Transcript
The treasurer’s role is to ensure the organization is resilient and able to survive in the face of multiple, simultaneous threats regardless of whether the events can be predicted or not. The key steps to leading to a resilient treasury include awareness and visibility to your liquidity and to threats to liquidity. Also key is preparation, the ability to analyze your risk, to act and have an action plan in place.
Our three panelists today exemplify the keys to treasury resiliency in that they’ve worked to understand their risks, whether it be energy prices, FX and/or credit. And they’ve employed tools to help them illustrate the possible outcomes and aid in managing the risks.
Without further ado, I’ll introduce our Bronze Award winner, and that is George Zinn of Microsoft, corporate vice president and treasurer. George is responsible for investing and managing over $50 billion of Microsoft’s corporate assets as of June 30, 2009. He leads a group that manages the company’s worldwide financial and corporate risk, investment portfolio, strategic portfolio, foreign exchange, corporate and structured project finance and dilution management. George joined Microsoft in 1996 as a financial analyst.
In 2009 he was recognized by Treasury & Risk magazine as one of the 100 most influential people in finance. I think I was number 101. Under his leadership the treasury function has been recognized with the Pinnacle Award from the Association for Financial Professionals and also Treasury Today’s Adam Smith award for best practice and innovation. George holds a B.A. in economics and environmental studies with a minor in romance languages from Bowdoin College. He earned his MBA from the University of Washington while working at Microsoft. Congratulations, George.
George Zinn, corporate vice president and treasurer, Microsoft Corp.: Thank you, and thank you Treasury & Risk magazine. I’m accepting this award on behalf of our FX team. That’s very important to me to have you understand that this award is part of a team effort, he entire 104-person treasury team that is part of the risk group working in treasury at Microsoft. I grew up there, so I know every single one of them.
They helped Microsoft navigate the credit crisis over the last year, this team. And not only with the absence of a negative, but I would also add with a number of accomplishments that you’re going to hear about over the next two days from the table. Please stand up and be recognized, Bonita, Christy, Chad and Laurie, and you’ll meet Bertan tomorrow.
Before you think that we are overstaffed with 104, let me explain. First of all, the reason that I’m here is we’re very conscious of expenses, and I’m accepting on behalf of the FX team because we could not bring everyone here this year. The truth of the matter is that 80% of the headcount in treasury is worldwide performing the credit and collections function for us. And so really we have a relatively small staff performing a number of duties. If you’ll take a look at our earnings, you’ll see that we are actually doing a much better job than we’re known for in continuously taking out expenses. Our CFO says, “Never waste a good crisis.”
Financial Risk Management 2010 Transcript
But I’m here to talk FX. So close your eyes for a moment and recall about one year ago. We were in the largest financial market dislocation in certainly my career. In particular, the flight to safety that drove the U.S. dollar up in that window of time was probably one of the most notorious rallies, again, at least in the history of my career. People experienced this in very painful, different ways. One would be on the balance in the fast 52 numbers??? in their balance sheet transfer pricing, which meant essentially you would have huge negatives going through your P&L if you held a lot of FX assets. The team did not win an award for this, but they have certainly won an award from me for keeping me in my seat and hedging that exposure. I really want to thank them for that as well.
Now this particular award is related to what I would refer to as a price protection program. This program is somewhat Microsoft specific, but it really is strategic for the company. This is not something these gentlemen did as a routine in their day job. They were looking at how to add value to the corporation.
Our channel is set up in a fashion where most people who purchase our software are purchasing it from our partner ecosystems. We have distributors, and as you may know if you have purchased a PC, you will typically go to an original equipment manufacturer or a retail store. We have a number of partners that we sell through. Particularly in emerging markets where currency plays a big role in the pricing, we leave that to our partners. And they pay us in dollars for the most part.
Now what happens is that in a number of situations if they purchase product from us and there’s a huge move in the currency, they end up actually being underwater before they are able to collect the money from the end customer, in a huge adverse move. Now admittedly this can be symmetrical as well. But our distributors are not in the business of foreign exchange. They’re really in the business of selling our product on our behalf. And the truth of the matter is if you want to help your partner -- which is exactly what this team did -- you want to take their minds off of this foreign exchange business. As you might imagine, since we are in every country in the world, there are a myriad of partners out there. Some are very small and some that you would know are very large, –including Dell and HP..
Now, take the backdrop of last year. They worked with these partners previously, and they said, “Look, we know that you cannot do hedging as efficiently as we can. So we’ll do that for you, and then at the end of the day what we will do is we will true up effectively what you owe us once you get paid from the end customer and you’re paying us.”
At the beginning of the transaction, the team hedges it. We put it on our balance sheet. Now that seems like the difference between the partner taking the risk and the bank taking the risk is effectively us all transferring it. We’re only taking partner credit risk. We already have partner credit risk. We want our partners to stay in business. The worst thing that can happen in our business model is for us to end up not being in the market. We want to be in the market all the time. We want to always be selling.
As a result, if we can keep our partners in business and particularly in the worst economic times, we will come out stronger on the other side. And we would have gained market share when other partners for other competitors have gone out of business as a result of not having this insurance, or protection.
At the end of the day, the operation is effectively that we hedge, and then as you can see, we true up on the back end wherever the currency has moved. It’s symmetrical. It will be either a credit or a debit on the invoice to the customer.
We can do this because it’s the whole treasury team working together. This is credit and collections performing this back-end function while the FX team upstream is performing the actual hedging function. The most powerful thing about this program over the last year was when our partners came to us. This was a service we had been providing, but during that crisis they were doing this on their own behalf-- larger partners were not -- they were shut out of the capital markets. Everyone was aware of what was going on over that period of time, and I think people all have their own vignettes. But these people, literally either through usurious rates or not, were told to go away. They could not access the capital markets and hedge anymore. So we stepped in and performed that function over the last 12 months for these larger organizations that actually had been doing this all along for themselves originally.
Financial Risk Management 2010 Transcript
As a result of that, we estimate that we have paid out on our hedges more than $30 million over the last year to our partners. The second thing is that more importantly we kept them in business, and as a result of that we believe that we have gained share over the last year as a result of this program. So with that I want to thank the team and look forward to the rest of the conference.
Moderator David Stowe: Thanks again, George. Congratulations. You definitely defy the definition of lean treasury staffs today, but I know it’s a different situation. What I think George is saying is that not only did they partner with the business outside of treasury, treasury acted as a consultant to find innovative solutions to a bigger issue. To reiterate, we are all in this together. Between suppliers and our customers, the point is the risk has to be managed. Who’s going to manage it? In this case, Microsoft had the infrastructure and the expertise to manage it better, and it led to an improvement in their overall business.
I would now like to introduce the Silver Award winner. It is Adhi Kesarla, senior treasury manager, corporate finance at Google Inc.
Adhi Kesarla: Thanks, David. As part of Google treasury, I work in the corporate finance team. It’s a very small team. My colleague Jessica Traveler is here. Together we handle corporate finance issues, basically evaluation and cash forecasting. We’ve gotten into energy risk management recently for the past year. And we are also looking into renewable energy investments. So let me get into the subject that I am going to talk about today.
Many of you know that last year we had a run-up in oil prices, and this resulted in power prices in Europe reaching multiyear highs. Google has data centers in Europe so our power prices were impacted very heavily. So we contacted our treasury data center team and had discussions as to how we can help them manage this exposure. One of the things that came out of this discussion was to help them purchase power for this data center in Europe where the price paid for power would equal the market average or lower.
In the U.S., power markets are regulated, whereas in Europe power markets are deregulated and so we have to buy power at quoted prices. This poses a challenge in that you have to be able to hedge the risk The fall contracts were primarily from fear prices. Our objective was to get prices that were equal to the market average or lower. We could look at the fundamental analysis of where the fall prices were going to head, or we could do a technical analysis, looking at historical prices and seeing how we could deal with the objective. Given that we deal with a market average, we felt that the technical approach was much better, since we didn’t have any political expertise in forecasting power prices at that time.
The energy company that Google had a relationship with offered Google to fix next year’s price based on an average of 10 power contracts that Google could sell during the year. For example, if you had to buy power for 2010, we had all of 2009 to select 10 power contracts during the year, and an average of those power contracts would be your price that you paid for next year.
The questions that the team faced was, how do you pick, and when do you choose this power contract, because the price of the power contract changes everyday. And do you think today’s price is good, or do you wait until tomorrow or next week? So treasury decided, let’s develop a strategy to help them decide when to buy power.
One of the simplest things that we could do was spread the power purchase throughout the year, very similar to dollar cost averaging. If you spread 10 purchases through the year, each period would be over 36 days. But then we said, how can we improve on this? Is there something we can do to time the purchase that would give us a better price? So we built a statistical tool to signal when is a good time to buy. The tool used two parameters in a moving average and a standard deviation, and this was based on most recent power prices. Using this tool, there has been a sustained shift in power prices either in an upward direction or a downward direction.
Financial Risk Management 2010 Transcript
Then based on that, we get a signal: “Hey, today’s a good time to buy.” So depending on where the power is trading in relation to the moving average and standard deviation, you could have five different scenarios. If it is trading within one standard deviation -- we just bought at the end of the last 36-day period. If the power is trading anywhere between three standard deviation and one standard deviation we bought one unit. And if it was trading beyond less than three standard deviations then we had a chance to buy units. It means two clicks. The reason was that power prices had fallen significantly, so we should take advantage of that.
At the same time, if it was above three standard deviations, we had an option to skip and wait until the power prices came down. So this was our strategy. We said, “Why don’t we test this to see if it makes sense,” and so we tested it on calendar year 2009 data that was eligible for our data months, and the results were pretty good. But then we wanted to test for longer periods. So we chose a German company and data, which was eligible for 10 years. And the results again turned out to be good in the sense that the strategies that we employed brought our data close to the market average.
In addition to that, we conducted tests on similar data. We looked at increasing scenarios, decreasing scenarios, cyclical scenarios, and then we tested the tools to see if the results were similar. What we found was that in many instances the click was within the market off 1 point for euros and in some cases it was even better.
But the question that we also wanted to answer was, what if we clicked on a random date? Would we end up with the same number? Would it be any different? So we ran multiple simulations, and each simulation was over 5,000 runs. We computed what the average would be if you had clicked on a random date during each of those 10 periods. Not surprisingly, the random click average and the market average were almost the same. But what was different was there was a significant spread. And the spread was anywhere from 6 to 9 euros between the mean and the max.
Basically, what it was telling us was, “Yeah, you can click it randomly, but you can end up paying a high price if you clicked at the wrong time.” What the click strategy does is to raise the spread of the distribution from 6 to 9 to something more like a plus or minus half-euro. So you’re always close to the market average.
In another significant take out one day of certain price -- you know when you have downward price movement,(you can actually end up beating the market average substantially. Based on those reasons, we were able to communicate to the data center team “Let’s use this strategy to buy power for calendar year 2010.” with the strategy we finished execution of the clicks and the price that we have is substantially lower.
One of the things that we asked during the discussion was: What if we increased the number of clicks from 10 to 20 or 30? Would that get a better price? Yes, it does, but it costs us extra money and so we didn’t pursue that fact.
The click strategy worked very well for us, and other than that there were a lot of benefits in terms of developing a good partnership with our data center team. We are working with them on how to manage their power exposure across parts of the world.
Moderator David Stowe: Are you surprised you have some of the most high-tech companies in the world and you get high-tech solutions to a very high-tech problem? I am sure you are all going to digest that in your head. What Google has done is brought a tool to a situation to aid in their decision-making process. In this case, when to buy or when to execute their hedging in energy prices in their European data centers. One of the biggest pitfalls in risk management is really not having objectives and not acting or not knowing when to buy. So they’re making headway in this endeavor with a decision tool.
Financial Risk Management 2010 Transcript
Our next and final award for this category is obviously the Gold Award winner, and it goes to Dave Smith of Adobe Systems. Dave is director of financial risk management at Adobe.
Dave Smith: Good afternoon. This work is the result of a group of people at Adobe, and unfortunately I was the only one who was able to make the trip. But Edmond Tang (and John LeTourneau contributed significantly to this. You may all be familiar with Adobe through great products like PhotoShop and PDF, and our motto is that Adobe revolutionizes how the world engages with ideas and information. We just bought a $1.8 billion company in a deal that closed Friday and day one was Monday. So my slides are a little stale. I apologize for that.
David mentioned that it’s important that organizations understand the value of financial risk management. I don’t know what your experiences have been, but there has been a steady educational process going on at Adobe, ,ot necessarily within treasury, but outside of the organization as a result of this last financial crisis.
The first time I presented any work on counterparty credit risk the only feedback I received outside of treasury was, “It looks good.” I started to get a whole lot more questions over the past two years. So the goals of this project were, first, we wanted to quantify all the counterparty credit risk within the company. Kind of a lofty goal. We wanted to create one integrated reporting package that aggregated our risks to counterparties across very dissimilar transactions.
Foreign exchange derivatives, discounted share purchase contracts, these are contracts where a financial institution agrees to repurchase your shares. You pay the money up front. You get a guaranteed discount for the volume weighted average price and you receive shares over time. On larger contracts, a lot of credit risk is being extended initially on the contract as opposed to a foreign exchange derivative where the credit risk is going to be exposed at delivery. At the end, banking deposits, insurance contracts and accounts receivable.
We wanted one measure of credit risk that could be consistently applied across all transactions. And we wanted to develop a policy to govern how we would model and allocate counterparty credit risk. Credit risk has been around forever, right? So how is it different for derivatives than it is for fixed-income transactions where there has been plenty of research done? Well, the cash flows from a bond are a certainty. For the floating-rate transaction, your coupon will be moving around but the principal is a NOMA certainty most of the time.
What’s different for a derivative is the amount of default risk is uncertain. You have a current exposure, which is today’s mark-to-market value, or your replacement cost of the derivative. And you have a potential exposure, which is what your mark-to-market value of that derivative could be in the future, and a lot more difficult to get at. You guys probably price your derivatives at least on a monthly basis if not more frequently so you know what your current exposure is.
But for potential exposure you need a simulation engine to understand that. You want something that’s going to simulate the underlying to know what the risk is there. The second thing about derivatives is that the timing of default matters. Your current exposure is going to move over time as the underlying moves so when the default happens is really going to matter. If it happens right after the month end then it’s pretty close to your current exposure. If it happens six months from now it can be a completely different credit profile.
Financial Risk Management 2010 Transcript
So your current exposure is not a great metric of what your risk could be. And current exposures aren’t something that get aggregated really well because if you have two transactions with the same mark-to-market value but one of them is a year longer in maturity, you have a very different risk profile there but you could have the same current exposure. The same is true if you’ve got two derivatives that have the same market value but very different volatilities. Then you could possibly have a much larger credit exposure with the more volatile asset. We need to look beyond just current exposures. What we need to look at is the expected loss due to default.
I’m a bit of a quant geek so I’m very excited that there’s some Greek up here on the board. Hope you guys don’t mind. There’s no quiz. So the first step to this, we simulate the stochastic path of the underlying, the value of the derivative at each time step. With all of our stuff we use weekly intervals. Determine the expected loss due to default at each time step, conditional on the firm not defaulting to that point. That’s basically solving first survival rate.
Multiply this conditional expected loss by the value of the derivative at each time step and then sum these amounts to get an expected loss for the contract under this particular situation. I’m going to show you some pretty graphs in a little bit that’ll maybe make this clear. But I want to mention that we use credit default swap spreads as the market’s expectation for expected loss, and I’ve been working on default research for a long time.
I think credit default swap spreads are the Holy Grail in terms of market-based indicator of default. They’re forward looking, unlike a rating agency and it’s market based, unlike a rating agency. Some of the criticisms of credit default swaps are that they move around too much. One week something’s got a credit default swap of 200 basis points, and next week it’s 600. We have had some banks that have been around for a hundred years and disappeared inside a week. I want my risk metric to be that volatile as well so I know exactly what the market thinks is going to happen.
We have four different levels of reporting that we put together: the individual firm reports; reports on counterparties aggregated across products for specific counterparty; reports within a transaction type, like foreign exchange; and then an attempt to look at an aggregated credit risk, or an aggregated counterparty credit risk.
So here’s what we have for our individual firm reports. We have ratings, short-term and long-term ratings. We found them to be really important not just the ratings, but the ratings outlook. We get automatic feeds on when the outlook was last updated, because that makes a big difference in how comfortable we are with that rating. If it was six months ago and we’re concerned about a credit, that rating doesn’t mean anything to me because it’s so stale. If it was two days ago I’m going to feel a little more comfortable that somebody there has paid attention recently.
We also have historical credit default swaps over time and the share price. I’m not a big fan of looking at the share price as the only metric for credit risk because the difference is in capital structure. If you have a firm that cuts its dividend and its share price gets tanked, their credit spread should tighten on that news.
We ran these reports on a daily basis, and sometimes even more frequently than that when things were really scary last summer. We would look at how the bonds were trading because you can only get CDS spreads after the close of the day. These are all reports by exposure type.
Financial Risk Management 2010 Transcript
For instance, this is our FX banks. We would produce these on a less frequent basis. We wouldn’t do this daily, but we would distribute it to our foreign exchange traders. They are sorted based on their CDS spread, with the biggest credit risk at the top. So hopefully that would stick out. We also used this as a metric to determine who we should get in place with now to start trading, what banks we would want to consider if we were concerned about our portfolios.
This is the prettiest of my graphs. This is all of our exposures with one bank at a snapshot in time. You can see on the bottom what transactions we have coming over time, and the purple line is the 50th percentile of our exposure with them. And you can see more extreme events. It’s likely we could have credit risk extended there. We’ve used this whenever we’re concerned with a bank and wanted to know what we had in place with them. Should we stop trading with this bank? Maybe we have a very large exposure, but it may be due to real short forwards. We would be more comfortable once those roll off that we could continue trading if their credit level doesn’t get too high.
And this is our attempt to aggregate what we could aggregate. We’ve got our whole slew of information. We’re big fans in Adobe treasury of extremely small font so I apologize for this. So we’ve got all the ratings metrics here on the banks. We’re trying to aggregate our current exposure and our expected loss for our foreign exchange contracts, our service purchase derivatives and our cash deposits.
What we use this for primarily is to look at it to see what scares us. There’s a lot of false precision in this. I’m modeling. I’m simulating the underlying. I think some of these reports even have things carried out to the dollar and cent level, and that’s somewhat ridiculous because this is not a precise science at all. But we use this to look at one quarter’s report versus the other and ask, what stands out? What scares me? What has changed in the exposure time from one period to the next? And I guess that’s it.
I wanted to mention one more thing. Our goal initially was to come up with a policy and say, “Here it is. Here’s our counterparty credit risk policy. We’re good for the next 10 years.” And we couldn’t do that. We couldn’t get there. If we had finished the policy in 2007 as if it were a fixed-income investment policy and we had limited everything based on expected loss, we could not have traded with many of our counterparties. I guess there may have been a few French banks, but other than that we couldn’t have traded and done our normal treasury operations throughout 2008.
The flip side of that is, had we finished this policy in 2008 so that we could actually trade, I think those risk limits would be excessive today. So this is not something you can quantify and fall asleep at the switch with. It’s something you need to monitor and look at going forward. Quarterly, we review these reports and make sure there’s a committee looking at it to see how comfortable we are. Thank you.
Moderator David Stowe: Thanks, Dave, and congratulations again. I think it’s obvious this is not your father’s risk management world anymore, to borrow a phrase. Obviously they bring very high-tech solutions to a very formidable problem. With the Adobe situation, I think that’s one of the most timely models out there. When we’re looking at the potential to standardize the over-the-counter derivatives, especially in the corporate world, which has a lot of downsides, credit will be one of those issues. I came from a formerly bankrupt airline. There’s a few of them out there, so I’ll let you figure out which one. We had obvious credit issues on our side and were therefore difficult to trade with. We were managing against margin risk in trying to forecast that at all times, so I can empathize with what Dave is going through.
Financial Risk Management 2010 Transcript
Questions and Answers:
Q: Hi, this question is for George on the FX hedging for your partners. I am curious as to what the uptake level was. Did the majority of partners in emerging markets participate in this? To me, it sounds like it would be a very manual challenge if you had a very large partner uptake. How did you manage that process?
George Zinn, Microsoft: As you might suspect, the uptake increased dramatically subsequent to September and October of last year. And of course it has tailed off again now. We were concerned that we would be overwhelmed, and we were not. So we were able to actually manage quite well. But that was a concern that we didn’t have to deal with.
Q: This question is for George as well. Actually, I have a multiple question. The key issue for us in the hedging business is to get the exposure data right. I’m imagining that if there are not exclusive distributors, it must be a challenge to share information. So how do you get that part right? And second, what are you hedging for them -- sales, pretax or cash flow? The other question I have is that it’s always easy for people to take the FX gain instead of the FX loss. So whether you have losses that you have to pass on, is there any difficulty there? And the last question is, what kind of instruments are you using in the hedging program? George Zinn: To answer the first question, the way we get the exposures and part of the reason that we weren’t overwhelmed is we were working through our subsidiaries. Our subsidiaries on the ground were aggregating those numbers for us. So that made it far simpler on the FX team. They managed that process. If you think about our partner ecosystem, which is over 10,000 partners, we only have a few hundred subsidiaries. So it’s a little bit easier in that respect.
The second part on the sales front? Obviously, we were getting the information and the exposures from the subsidiaries, and they were custom designing for the partner. So whatever the partner wanted to hedge. Some partners were not hedging, for example, 100% of their exposure. It was à la carte, I guess is the right way to say it, for the partners. It was across the board cash flow, portions hedging, 50% of pretax sales. So it would just depend on what the partner wanted to do, not what we wanted to hedge.
As for the losses, the way that we dealt with that is the fact that the partner didn’t experience them, and this was very important last year for the partners. They didn’t experience the loss until the maturity of the contract, and so when the contract matured, what we would do is send off an invoice at that time, which would effectively change or currency adjust the price to the partner.
To the extent the uptake decreases as the dollar weakens here, to answer your question about how they deal with that, it seems that crises fade from people’s minds pretty quickly. So it is interesting to watch that behavior.
And then the last is, in certain countries we’re actually not able to offer this program, which is where I think you are going with that. Because in certain countries there are nondeliverable forwards that we will use, which are the most exotic. But primarily just forwards. We’re not using any options in this situation because we believe that that is a little bit of a different service that we would be providing.
Q: This is a question for all three of you, but really more directed toward Google and Adobe because of the complex strategy that you are employing. Do you see this as purely a hedge, or are you actually trying to create a profit center out of your strategies?
Adhi Kesarla, Google: We actually see it purely as a hedge, and in the process if we can actually save some money by getting a better strategy, then that’s great. But it’s purely a hedge. It’s not a profit center.
Dave Smith, Adobe: Yeah, we’re not a profit center, either. We’re just trying to make sure we have no uncomfortable conversations with the board, that’s all.
Q: This question is for Dave. On the simulation engine that you used to model the default risk, is that an in-house capability or do you use a third party to assist you with that? And the second piece is, do you also roll up your customer AR exposures?
Dave Smith, Adobe: The first part is, no, we do all of our programming in MATLAB. We have got several licenses. It’s a programming language that we use for that, and it’s well suited to do that kind of simulation stuff. –But I’ve been programming it for about 10 years, so if I’ve got a hammer, everything I see looks like a nail. That’s why we’re using that.
With respect to our customers’ AR exposures, unfortunately the limitation is that our distributors, for instance, are our largest customers. And they don’t have publicly traded debt outstanding so they don’t have credit default swaps. We can’t really look at them on an apples to apples basis. We’re limited to doing traditional credit work. It’s difficult to incorporate that. We’ve made several different attempts to do it. But we’re not comfortable that it’s on the same playing field as it is for our financial derivatives.