Credit Risk Management 2010 Transcript

From Treasury & Risk's 15th Annual Alexander Hamilton Best Practices Awards

TOM DUGGAN: This is the last session of the day, so thank you so much for being here. This is the credit risk management category, and we'll be moderated by Chuck Kane. Chuck has more than 25 years of senior executive level experience, including being the chief finance officer at RSA Security, which is now part of EMC, Aspen Technology, Informatics Software, which is now part of IBM, and well as a role as president and CEO of Corechange, now part of OpenText.

Most recently, Chuck held several executive positions, including president and COO at One Laptop Per Child, a nonprofit organization founded at MIT that provides computing Internet access for students in the developing worlds. So now, I'd like to turn it over to Chuck for our last category. Great.

CHUCK KANE:  Thanks, Tom.

DUGGAN:  Thank you.

KANE:  So I have no slides ‑‑ they've saved the the best for last here, and there's a lot of empty seats. So I guess I should really go through this quickly. First, I want to thank all the folks that I've read all your entries. I was also a judge, and I appreciate reading all your entries instead of being at the beach this summer.

The book they sent us was about yea thick of all the write-ups, but they were incredibly insightful as to what companies are doing with technology and approaches and efficiencies, and these three finalists are a great indication of that.

I want to thank Treasury & Risk because I believe and I teach at MIT at Sloan ‑‑ I believe that treasury in general is the most underrated and unappreciated piece of the finance organization. I have come up through the ranks, initially in treasury, and I can tell you that with high tech companies that start out and get some momentum and become $100-$200 million type companies, usually there's one person in treasury doing it all.

There's no appreciation for the funds that can be saved through the efficiencies gained, and for the risks that can be avoided until something happens, and then it becomes real expensive. So I’m very pleased to be part of this program because it recognizes the treasury function for what it should be recognized for.

The other point I'd make before I introduce the speakers here is that credit risk is a double-edged sword. I'd like to make a point that it's a risk-return criteria and I'd like to do that by giving an example. I used to work for a company called Stratus Computer. It was a fault-tolerant minicomputer company back in the dark ages, and we were very successful in growing as a computer company.

And we had this client, this customer of ours in New Jersey, a small company that kept buying a lot of computers and asking for extended terms more and more and more each time they bought more computers. It was becoming quickly our biggest customer, but the risk involved in extending them credit and shipping those computers to them was a very iffy proposition because they had very little capital to back the organization.

So we decided to shut it down, finally. Now that company became AOL. We had shut down and lost that customer at the front end of what would have been an amazing customer for us long term. And we did that by careful determination of what was our risk return on that deal.

So it's not only about what you don't collect. It's also about what you give up by not taking that credit risk. And I want to make that very clear that that's why it's part of that ‑‑ the dynamic that you have to deal with in the treasury function.

And then, lastly before again, I introduce the folks here ‑‑ I'm with a nonprofit organization called One Laptop Per Child. We sell directly to governments -- so we sell computers to the government that gives them to the poorest children in developing countries in Africa and Latin America.

My credit risk is very simple. I do an LC or I don't do business. So unlike my easy task at this right now with governments, these folks have to face a lot more challenges and a lot more procedural checkpoints in order to take risks.

So with that, I will introduce the winners here. First, the Bronze goes to Omnicom Group, Maeve Robinson, who is the assistant treasurer of Omnicom. Maeve, congratulations. Go ahead.

MAEVE ROBINSON:  Thank you very much. Omnicom's our first time up here at the podium. So we're thrilled to be participating and thank you for the recognition. Just a little bit about Omnicom and I try to keep my slides to a minimum so I apologize ‑‑ I don't have a slide on this but Omnicom is not as well-known as some of our operations.

We operate a very decentralized organization, about 60,000 employees in about 900 locations, about 100 countries around the world. Any of you Don Draper fans, you probably heard us mentioned throughout those episodes ‑‑ we're an advertising and corporate communications company. We own companies like BBDO, DDB Needham, TBWA/Chiat/Day out on the West Coast. But we're all over the place. And Omnicom isn't really a household name, but you probably have heard of our companies and seen a lot of our work.

But credit risk management is critical to our operations, and that's why we developed this credit program. Here's again by way of background our balance sheet, very summary balance sheet. You can see we have $5.2 billion approximately -- this is as of June -- AR, which is our second biggest asset aside from goodwill.

Our revenue, on a 12-month basis, is about $12 billion. So on the surface you might say, ‘They're doing a terrible job collecting.’ But because we are this ad agency where a true agency relationship with many of our clients and much of our business, you have to really look at our cash flow because we're billing and collecting from our clients, but we're paying most of that out in ‑‑ to our vendors, you know, the media companies and the production companies around the world.

So our cash flow, our turnover, is more like $60 billion. So we really only have about 30-to-40-day DBO or days payable, or days billing outstanding, as we call it, vs. what you might see if you were just to measure our $5 billion of AR against our revenue of 12 ‑‑ it's more like six months but it's not that bad ‑‑ it's about a month.

But it is critical, as you can see, based upon these numbers, that we manage our credit risk as effectively as possible because one bad debt will have a significant impact. It'll drop right through to our bottom line. And it wasn't really ‑‑ it had nothing to do with our revenue. It's geometrically huge compared to our revenue.

So we developed a three-pronged program to address our credit risk management. Part of that is a working capital program ‑‑ we're looking at AR. Part of it is credit analysis. And the third part is credit insurance.

We had systems that addressed these three items before we pulled them all together. Each of those systems was effective for the time we were operating in then. But now, since 2008-2009, it's a whole new world out there. All of these had to be redeveloped and were redeveloped and are now global, and they're automated. We put them all up on the Web. They're accessible from all around the world on an automated straight-through process basis. And we now have visibility all the way down through the 900 operations to the client point of contact in each case.

So I want to talk just briefly about each of the three components. The working capital program has timely and continuous monitoring of our accounts receivable on pretty much a weekly basis with all of our operations around the world.

We collect data once a month. It's all collected automatically out of their ledger systems. We hold conference calls with all of our operations around the world every month, at least once a month, to review the status of their AR. We focus mostly on our bigger items ‑‑ anything greater than $100,000 or that are 60 days old.

We also collect every week the cash that's collected around the world vs. those AR, so we have a weekly update of exactly where our accounts receivable stands all the way down again to the client level, the client point of contact at every location around the world. We port all this up through senior management, and now we're able to aggregate our exposures by client all around the world. We didn't have that before in our old program. And the working capital program is integrated with the other two pieces of our program.

Just a couple of screen shots ‑‑ I apologize it's so small, but this is the look and feel of the programs that we've developed. All of our operations again report in all this data on a monthly basis, on a weekly basis for some of it, and again, it's all the way down to the aging by client by location around the world. But then again, it all gets aggregated up.

The second piece is our credit analysis tool. This is probably the place where we were lacking the most prior to this program. We were pretty much relying on all of our finance directors and controllers around the world to be doing this on their own. But we realized that that was probably not very good to rely on that group to be doing it. So we developed this common framework for everybody to use. It was developed for the local finance people, so it incorporates all the regional and local practices that might be involved in evaluating client credit. We look at credit at both the gross and the net exposure basis. The gross exposure would be typically our AR, our work in progress, and also what's a little bit unique to our industry, our non-cancellable commitments.

So we do go in advance in the beginning at some point in the year. They will contract to purchase media and other services on behalf of our clients, which are non-cancellable at some point. So that plays into our gross exposure calculation. With the workflow product embedded in this tool, these requests for credit limits get processed up through the organization, depending upon the level requested. It'll flow up through various levels of the local organization, the regional organization, network, and all the way up to the corporate organization. And, of course, we expect these credit evaluations to be periodically updated. We don't expect to see them updated every month but as often as necessary, especially if there's a significant credit event.

And, again, we can link ‑‑ this is all linked into the working capital. And the next thing I'm going to talk about is a credit insurance database. What I like is when we get our updates out of our working capital, whether it's the monthly updates for the month-end balances or the weekly updates for the cash collections, that gets fed automatically into the credit analysis tool so if that triggers an issue because the credit limit is not enough, because their AR exceeds our credit limit, there's a whole workflow on that, and everybody is required to then go back and either get an update on their credit limit or get to collect the receivables so they don't have that issue anymore.

Again, a couple of screen shots for our credit analysis tool, a similar look and feel to it, by agency, by location, by client. There are several pieces of data we collect in order to do the complete credit evaluation, including financial statement information on our clients, our rating agency information of our clients, and then what our expected exposures are going to be. And then the tool itself calculates the gross and net exposures, and a limit is proposed locally.

We also, separate from our credit evaluation tool, we established on our intranet the ‘credit evaluation place’ we call it, and we just keep current events published up there relating to any of our clients that's out there in, you know, the public domain. It's pretty much automatically fed from various news sources and if you have any questions on your clients, you can go here, and see if there's anything published or readily available.

The third leg is our credit insurance database. Again, prior to pulling this altogether in one place, we had credit insurance that was purchased and negotiated locally. Most of this is outside of the U.S. There's very little of that in the U.S. But in many Asian and European regions, it's very common to have credit insurance in our industry.

But again, all of that was very localized. We didn't have visibility at the top, and we really didn't have a handle on what we had insured and what we didn't have insured. So with our risk managers, we pulled together and centralized the procurement of credit insurance into a single global program. It is administered locally, however, to allow for local customs and practices. We identified and cleared up gaps in coverage. We were able to reduce our premiums substantially, and the carrier that we chose keeps all of its data -- has a very robust portal, where it keeps, maintains all of its data on our insurance and our clients.

So we're able to have access to that and we pull that into our own systems on a daily basis ‑‑ again, on a client-by-client operation, by our operation basis. So we're able to pull together the three areas in one comprehensive database. We have all of our working capital with AR. We have our credit evaluation tool, which is our evaluation of the credit risk of each client. And we have the related insurance, where that exists.

So where we ended up with was working capital, credit analysis tool, and our credit insurance database and program all linked together in one comprehensive client-credit information system. Thank you.

KANE:  So our Silver Award goes to Microsoft Corp., and here to accept from Microsoft is Stephanie Alston, and Stephanie is a senior credit manager. Stephanie, I understand that that good earnings report yesterday was due in large part by our credit management efforts in Microsoft. Is that true? Of course, it's true, okay. Why don't you come up? Congratulations.

STEPHANIE ALSTON: Well, as Chuck mentioned it, it is all treasury, of course, and credit management, so we were very excited to see the earnings yesterday as well. So things are looking great. Today, I'm going to focus mainly on our accounts receivable, specifically in the Microsoft advertising stream in EMEA.

So at the start of 2009, we were really facing a significant challenge, where our 60 days or over-60-day AR was at 22% of total AR, and by June it had escalated to a high of about 29%. In addition to that, we had disputes that were escalating.

We had $21 million ‑‑ a high of $21 million in dispute. In the advertising business, a lot of times, the pricing, contract issues, agencies tend to have several different entities, so if the business gets booked under the incorrect entity, you would have a dispute there as well. So we saw those continue to increase.

In addition, and specifically in the EMEA market, we have a lot of high-volume, low-value accounts that we were really managing manually. So taking into consideration all the languages that we were supporting, the different cultures, and the high volume of these low-value accounts, we needed to find a way to really be able to manage this better.

So the questions we presented ourselves with were, you know, how do we reduce our past-due AR and collect the revenue that we were generating? How do we further enable the revenue, and then mitigate the risk that we have without impacting customer satisfaction? So that's really where ‑‑ those were the questions we asked ourselves before we started, and we really took a multi-pronged approach to fixing this.

After doing a little bit of analysis, we realized that there wasn't one tool or one system that was going to get us to where we needed to be at the end of the day. So we started with a couple of different reporting tools. The first one was our risk reporting, which really enabled us to view where our exposure was. So we pulled the parent company data for each advertiser and agency, whether they were publicly traded or private and the subsector and sector that each one of the customers was in. And it really enabled us to see where our credit risk was in the risk profile of each customer.

The second thing we did was we really needed to engage our sales team because without the support of our sales team, we really can't continue to collect on the AR and engage our customers. So we developed a second report, which really we wanted to put in a format that sales could understand and quickly action. So we detailed AR by invoice, as well as disputes. We also included all of the customers that had been placed on credit holds and the sales allowance reserve that we had for each country.

We then took this report and put it on to SharePoint, where they could easily access and update the information. So essentially we built a dashboard for sales and credit and collections to be able to utilize and communicate quickly and effectively.

The third thing we did was, which was vital to our success, was redesign our dispute resolution process. So if sales is responsible for inputting the disputes or, excuse me, the campaigns into the system, and if they're inputting incorrectly, as I mentioned, it raises a dispute with the customers. So in order to fix this, we took the ownership from what we consider the account resolution team at the time and put it back on the sales organization.

We also implemented in a lot of timeframe that they had to resolve these disputes, which really helped sales management to focus on the root cause of the disputes, and to mitigate a dispute from ever occurring, which was great.

So we love our sales teams. We partner well with them but we knew to keep them engaged and accountable for this dispute resolution process, we needed to take it a step further. So we worked with our finance teams to revise our sales allowance methodology, which historically we were under-reserving against revenue for the disputes. So we took ‑‑ the new calculation was really a percent of the disputes and depending on their age, that percent either was smaller or higher.

That ‑‑ it was then ‑‑ it's done monthly and it's booked against revenue, so it really encouraged the sales team to be accountable for resolving their disputes. The big thing that we did that I think really enabled us to be able to focus more on our top-tier, more strategic and complex customers was we internally built a system on Microsoft Access that allows us to take customers, really manage the day-to-day systematically, and we can take a customer entirely through the dunning process all the way to credit hold with just the click of a button.

We can also manage all of our customer communications, all of our contact information for our customers, and where they are in the dunning cycle through this tool. So again, it really just enabled us to manage that bottom, actually the 80% of our customer base that is really high volume.

Then the last thing we did in conjunction with that was to really focus on our credit-hold process, which leading up to this point, hadn't been as adhered to as it should have. So we revamped the process, spent a lot of time educating sales as to what that process was, as well as our customers, and once that was in place, we really saw a significant improvement in our past-due ARs.

So taking all these things together, since we've implemented all of these, the EMEA region has now become basically the leading region for metrics, consistency, and customer satisfaction. We've reduced our 60-plus AR by over 80% at this point, and our disputes have decreased by 90%.

We're able to systematically manage our high-volume, low-value accounts, and ultimately and probably most importantly, we have very high customer satisfaction, and our sales teams are happy. So, thank you.

KANE: Thank you, Stephanie. It's great to hear when companies use their own technology in delivering things, although I imagine it's pretty easy for Microsoft to do that with a lot of things.

Our Gold Award goes to Cisco Systems, and Greg. So Greg, before I give you the mike here, that story I told about Stratus Computer, our head of marketing at that time was a gentleman by the name of John Morgridge, and John ultimately became the chairman, long-time chairman, of Cisco Systems, one of the original founders. So I can blame John for shutting off AOL in this endeavor. Greg Bromberger ‑‑

GREG BROMBERGER: Thanks a lot. I appreciate it. Thank you all. Yesterday I was keeping you from the open bar and today it's lunch. So I'm sympathetic to that fact and we'll get right into this. But even though this is a credit category, I mean, this is another really good story about collaboration at Cisco.

You know, our prior state was we had this massive balance-sheet cash-investments portfolio that was managed by treasury that contained a lot of credit risks, but we realized that this was only kind of part of the overall exposure to the company. There were other groups, namely, Cisco Capital and Trade Credit, that were also housing significant amounts of credit risks, and weren't having kind of a coordinated, aligned oversight of these three at the time disconnected portfolios.

So what we really endeavored to do was try and reconstitute the advisory group that we had in that singular focus over the treasury portfolio to really take a look at this idea of at least initially integrated credit risks.

So rather than just kind of overseeing and advising the portfolios, we really wanted to institute this idea of getting great alignment, as well as collaboration. And even though we initially started out with credit, which is what I'm going to talk about today, we've since expanded it to what are the overall integrated interest-rate risks across the portfolios. And now that we've got kind of credit and interest-rates risks being quantified and managed, what about the return aspect as well? How do we get that involved to make sure, as Charles mentioned before, that we're not just focused on one ‑‑ one side of the equation?

So in terms of scope of the project and what was

ultimately going to be combined and managed, you can see for a Cisco Capital group that largely consisted of leases, loans, and guarantees, trade credit would be receivables, and then the traditional treasury portfolio with our fixed-income securities, but then also layering in the credit exposures that we have from our cash counterparties for bank accounts, but also for our derivative exposure, and how do we get these disconnected asset classes all into one big portfolio that we can manage?

In terms of the challenges, I don't think that there are any surprises here, so I don't want to belabor any of this but in terms of the data, it was significant. We had 2,000 counterparties, if you will, and nearly 40,000 separate line items across these three portfolios that we needed to manage.

Therefore, systems and tools were hugely important. We ultimately went with KMV, which is a Moody solution that allows us to take market-driven views of our exposures, rather than relying on static credit ratings. So we use their expected to fall frequencies, the major output that comes from that tool.

The other important thing here was defining a risk appetite, right. We each had been doing kind of our own thing in each of our three worlds and having to come together and decide on what would be the common, you know, quantitative thresholds and metrics that we would use across the three portfolios. And we decided on largely Value-at-Risk and catastrophic risk thresholds that we ultimately set.

And then again, I know we're probably sick of hearing the word TelePresence and collaboration, but this was a project because the leaders of each of these groups were based in San Jose, Chicago for our credit exposure, and Atlanta for our Cisco Capital. So to get these three teams working together was a great example of how we could rely on the technology to make what historically would have been a difficult project actually achievable.

So once we kind of solved what the groundwork would be, we had to ‑‑ the next most important thing was, what are going to be the credit risk measures, and definitions that we were going to use? And I don't want to get into the quantitative weeds on this, but we did decide on these three metrics, you know: expected loss, which is greater than 50%; Value-at-Risk ‑‑ it's kind of your 1-in-20 twenty scenario that we're all familiar with; and then the catastrophic would be, you know, that one in a thousand year flood.

So in terms of the output now that we can all

view as a group, and I mentioned in my discussion yesterday, that this is something we report out to senior management on a quarterly cadence. You know we can view this consolidated portfolio now, which from a credit standpoint is, you know, $15-odd billion, and you know for the interest rate and kind of the total return is actually a $50 billion number. We can look at it by instrument, geography, industry, and actually implied credit rating, so that's not actually the agency's rating but what, based on these market-driven events, would imply that our holding should be rated.

And, again, the most important thing that we did is we set up these risk thresholds, so we can now look at this integrated portfolio against these important thresholds that we've set, which are basically linked to, you know, as a percentage of the overall portfolio, what the Value-at-Risk might be or catastrophic risk from any, you know, single exposure driver. And then as well, we decided on as a percent of our market capitalization, how much risk were we willing to take in our credit book and still be comfortable that we've got the right balance on.

We've also got other things that we look at ‑‑ we look at our top 20 exposures as they contribute to VaR and catastrophic risk as well, and then we also make sure that we don't have concentration risk, as you'd expect, right? We try to keep everything under a certain percentage of our overall portfolio.

This is just kind of our evolution. I don't want to get in the details of this but I do think the most important thing in the box up there is how we've made this transition from working in our own silos, which actually worked very well, you know, we were optimizing each of the individual functions, if you will, but we were kind of missing that bigger picture. So that fact that now we're working cross-functionally as a team, even the ad hoc things that we discuss with each other, I think, can make a difference on the overall portfolio. And again, all this is a vision and kind of the steps that we took to actually manage the credit component.

We're really excited now about layering on the interest-rate risk and the total return aspects of the integrated portfolio. So that was it ‑‑ I thank you all for the recognition and the award. And I'm done.

KANE:  Thanks, Greg. Before I start the question-and-answer period, I hope you people appreciate and acknowledge, one of the components that I've been seeing as a common theme in all these presentations is the integration effort and how technology has made things so integrated in real time. And it's an amazing advance and I've been out of treasury for some time now. The way that I would go after my DSO is put a bonus in front of my collectors. And now I look at these examples of how the companies are integrating their sales force with instant information on issues in the treasury area and also a way to enhance the bottom line of a company by managing these areas so efficiently.

I'd like to kick off the question, first of all, and this was always a pet peeve of mine. If, indeed, there are tough collections in the company, do you go back after your sales force, and do you take back your commissions ‑‑ any clawback? I love clawback.

ALSTON:  I can see that.

KANE:  Go ahead, Stephanie, you had mentioned in yours.

ALSTON: We'd love to.

KANE:  Yeah.

ALSTON:  We would love to do this. I mean, we couldn't do it from a commission perspective, but we could do it from the country's bottom line. So adding that additional sales allowance reserve really incentivized, maybe not the individual sales folks, but the country manager whose bottom line was, you know ‑‑

KANE:  And he'll take care of the individual sales guys.

ALSTON:  Was being affected and he was taking it to the individual sales guys, yeah, so ‑‑

KANE:  Maeve ‑‑

ROBINSON:  Well, we don't have sales force ‑‑ we have account managers.

KANE:  Yes.

ROBINSON:  And, yes, there's many of our better performing operations, the account managers are an integral part of their collections. In fact, many times they are the collectors ‑‑

KANE:  Uh-hum.

ROBINSON:  Because they're the ones that want that relationship with the client there in their offices that they could be there and actually pick up the checks and, you know, they have ‑‑ they understand better the relationship. So that's been around in our business for a long time and it does work very effectively ‑‑ yes.

KANE:  Right. Greg ‑‑

BROMBERGER:  The only thing I'd add is that we used to actually comp our sales force on bookings, not even sales. So the fact that we now even have made the step to where ‑‑

KANE:  I can book anything if that's the way you do it. Could I be a sales guy out there?

BROMBERGER:  It was good to be a salesman at Cisco in the ‘90s, I'll tell you that but, yeah, you know, we ‑‑ we basically ‑‑ similar to what Microsoft had talked about.

KANE:  Do the analysts still look at DSO as a major indicator of both quality of product and, you know, management of your balance sheet and stuff?

BROMBERGER:  They do ‑‑ I'll just quickly answer that ‑‑ that's still one of the key metrics that our CFO talks about with the Street and kind of that cash flow conversion cycle, actually on both sides of the balance sheet ‑‑ yeah.

KANE:  Yeah.

ALSTON:  Yeah.

KANE:  Same?

ROBINSON:  Yes.

KANE:  Great. How about some questions out in the audience? Do I have anybody?

QUESTION:  I have a question for Omnicom because you are a very decentralized organization. When you took some of the credit risk decisions to headquarters, did you find resistance from the different businesses around the world to that?

ROBINSON:  Well, we didn't really take too many decisions away from the local management because we do operate very decentralized, we are very entrepreneurial ‑‑ that's where the relationship is with the client and that's where we want the ownership to remain.

So it's really, it's really just an approval ‑‑ it's less of a responsibility taking than just making sure that the right levels understand the risks that the local operation is taking on. There has been an occasion where a level of credit exposure, for example, a credit limit was denied.

There aren't that many because, generally speaking, there's a good business case for what's being requested. But the responsibility stays at the local level, and we need it to be there to maximize our business and our operations and our client relationships.

QUESTION: Do any of you buy credit insurance and under what circumstances?

BROMBERGER:  We used to ‑‑ right, because that's when we were very kind of, I guess, selectively focused on the individual portfolios, especially for some of the concentrated happenings, and say, our trade accounts book. But since we've actually taken this integrated portfolio view, we get much greater diversification across the bigger portfolios. So we've been less ‑‑ less needing to actually do that recently.

ALSTON:  Yes, so within treasury, within Microsoft, we have a risk team that really looks at that and focuses on that; specifically within the Microsoft advertising stream, we do not do the credit insurance.

ROBINSON:  We have an extensive credit insurance program, primarily outside of the United States. It's very common in our industry, and it's very common in those regions, where it's very popular to have that as one way of, you know, offloading some of the risks associated to our business. And because of the dynamics of our agency relationship with our clients and the amount of pass-through costs, it's a very efficient way to manage that risk.

KANE:  I know this was probably covered in one of the other sessions, but those receivables and a lot of your business is international ‑‑ I'm assuming you hedge all of your exposures when you manage your receivables that way? Are they all hedged ‑‑ from currency standpoint, yes?

BROMBERGER:  Ours are.

KANE:  Yeah, totally hedged.

ALSTON:  Yes.

KANE:  Yes, okay. Your stock would have gone down if you said no ‑‑ you know that?

Are there any other questions out in the audience? Everyone wants lunch? Well, thank you, folks, for participating in this and I think there's going to be some final words said up here. Who's saying the final words ‑‑ are you leaving that to me? Thank you.

Page 3 of 9
Comments