Batterinram is correct. All the companies mentioned by him have outsourced their modeling work to India. Most of them have their captive centers and hire graduates and postgraduates (even CFAs and CAs) to do the work. Some of the BBs already having an operational center in India are JPM, Lehman, GS, HSBC, Fidelity, UBS, MS, Bank of America, Citi; even some hedge funds like DE Shaw have their captives in India. Apart from these, many more banks work with the third party outsourcing companies based in India, China, Latin America and CEE for their modeling and research work. I am workign at one of the third party outsourcing firms. The cost arbitrage is a major driver for outsourcing industry and the BBs are the ones taking maximum advantage. Outsourcing began with Business Process Outsourcing (customer care) and IT Outsourcing services and now it has moved to knowledge process outsourcing which includes outsource of services like market study, feasibility study (for companies and markets), financial models etc. Even many of the consultancy firms are outsourcing such work. This is a big market now. And due to the credit crunch most of the firms are opting for outsourcing as in order to achieve cost cutting.
batterinram Wrote: ------------------------------------------------------- > BigBean wrote: > “Maybe this is the future, but for now there > doesn’t seem to be the expertise, though I know a > London based company that is aiming to tap into > this market. I tried to convince my old company to > let me start up the Indian side, but they were too > old school for this.” > > This is already happening Goldman, Fidelity, JP > Morgan, HSBC have all outsourced modelling work to > India. Some small firms are starting to do this as > well. I guess it depends on the industry you work in. In my industry it is yet to happen - which is not to say it won’t.
numi Wrote: ------------------------------------------------------- > Good assumptions about a potential investment > drive the financial model, not the other way > around. Usually, good analysts have a gut feeling > about a potential investment based on their > fundamentals in advance of putting a model > together. A careful analysis of a company’s > fundamentals identify whether a business is a good > investment; the model answers the question of how > much you should pay for it. That being said, the > model is still absolutely essential in order to > pinpoint valuation. Numi: If you have time, can you please walk us through your approach in building a model for a company? That shall be much appreciated. Also, now that you are in PE, do you find a huge difference in modeling between research and PE?
Work that requires good IP protection is less likely to be outsourced, although it still could go to Canada or Ireland, as opposed to India and China.
gauravku Wrote: ------------------------------------------------------- > Batterinram is correct. All the companies > mentioned by him have outsourced their modeling > work to India. Most of them have their captive > centers and hire graduates and postgraduates (even > CFAs and CAs) to do the work. > > Some of the BBs already having an operational > center in India are JPM, Lehman, GS, HSBC, > Fidelity, UBS, MS, Bank of America, Citi; even > some hedge funds like DE Shaw have their captives > in India. > > Apart from these, many more banks work with the > third party outsourcing companies based in India, > China, Latin America and CEE for their modeling > and research work. I am workign at one of the > third party outsourcing firms. > Can you please go into more detail about the extent of your modeling responsibilities, and what you guys actually do as “outsourced modelers”? I have actually worked at not one, but two of the companies you mentioned above – and at least in my personal experiences, I felt that the modeling we outsourced overseas tended to be more mechanical in nature, i.e. setting up a model template, populating it with data from 10-Q and 10-K, and so forth. And while our counterparts overseas did a good job in this area, we (the analysts and associates in the US) typically retained responsibility for building the projections. In case you’re wondering why this is so, it’s because we – the analysts and associates in the US – are in the flow of information every day by being at work when the markets are open and constantly speaking with the traders, salespeople, and buy-siders. And what that means is that since we’re in the live loop of the markets the way that the folks overseas are not, it’d be easier for us to maintain and revise the projection assumptions ourselves, rather than having to communicate these assumptions overseas and having our peers there do it. Furthermore, and perhaps more importantly, we understood the companies far better than what our “outsourced modelers” did, and at the end of the day, it’s the assumptions and knowledge of a company’s fundamentals that drive the model, not the other way around. Anyway, I’d appreciate if you can clarify the scope of responsibilities that outsourced modeling entails. I think a lot of us have divided opinions as to what modeling really is – in my view, just setting up a model and pulling historical information is not modeling, whereas building the forward looking projections is. And it’s far more sensible for someone based in the US to do the latter, especially if we’re talking about a company in the US market, simply because they actually have the necessary sector knowledge to do so. If you guys can be more specific about what is being done overseas versus what is retained in the US, this could make for a more interesting discussion as opposed to just debating whether or not more finance work is being sent overseas (with the short answer to this question obviously being “yes”).
abacus Wrote: ------------------------------------------------------- > Numi: If you have time, can you please walk us > through your approach in building a model for a > company? That shall be much appreciated. Also, now > that you are in PE, do you find a huge difference > in modeling between research and PE? The best way to think about PE modeling is basically a combination of research and banking modeling. In research, you have primarily operating models – basically models that focus principally on the P&L, balance sheet, and cash flow, and building a financial representation of a company’s operations with as much granularity as is sensible. The idea of valuing a public security is essentially based on trading multiples, or a discounted valuation on its projected future cash flows. Research models don’t typically include assumptions about potential transactions, but they tend to be pretty detailed because the senior analyst is very knowledgeable about a sector in the way that the average banker isn’t, simply because that’s his or her specialty. On the banking side, you definitely look at a company’s operations as well – how you value a company doesn’t fundamentally change at all – but building up the different revenue and expense items is often less precise, and in fact, many bankers just get the operating projections from research reports anyway. Where the bankers come in is principally on the transaction side – their goal is to evaluate a wide range of potential transaction scenarios and sensitivities, and advise their client on the deal that makes not only the most sense for the management team’s objectives but to also garner the highest possible valuation for that entity, within reason. That’s the banker’s job – to help execute transactions, be it M&A, LBO, or whatever – and extract the most value they can so they get paid more. On the private equity side, I would say that having good transactional and research modeling skills are very important. It’s definitely more on the transaction side, particularly with LBO modeling, because we need to know what we are willing to pay for an entire company and also understand how IRR’s will change based on how much debt we can get or how we structure the transaction. But at the same time, because we’re on the buy-side (as opposed to the bankers who are on the sell-side), we are not looking for the most “optimistic” valuation a company – we want the valuation that comes closest to reality based on what our assumptions are. Thus, when we are engaged in the bidding process for a particular company, the bankers and management team of a potential acquisition target give us access to the data room, which is an electronic file containing a number of internal documents about the entity – these may include customer lists, monthly P&L’s, internal memos and projections, and so forth – basically all types of information that would not be available to the public. It’s amazing how little information is actually available to public investors till you’ve seen the private side. But the point is, since PE investors are in the business of buying companies, we basically want to know every little thing about a company and be able to anticipate every possible scenario or risk that can potentially affect the company. As a result, we not only care about building a thorough transactional model since that is what gives us an idea of how to structure the deal – but we certainly care about building a thorough operating model too, in order to best assess how the company will perform. And that’s where I personally feel my research modeling skills have come into play (not to mention that my research skills were far more developed than my M&A and LBO modeling skills, naturally because I came from the research side). Anyway, I guess this response is more detailed than I imagined it to be, but your question was a good one and one that was also raised by others in the past. Hopefully this helps.
Numi, that’s very enlightening. But I’m still a little foggy on how “transactional” modeling differs from “operational” modeling. Is it that you’re making a decision about whether to do pure equity, convertible debt, etc., and so you are running a number of future scenarios to figure out which has the best risk/reward? Or is it something else.
An operating model is primarily just forward projection’s about a company’s operating activities, i.e. income statement, balance sheet, and cash flow. It’s the type of thing that most CFA candidates/charterholders seem to be concerned about as public security analysts. Transaction models are M&A, LBO, restructuring, etc. There are a whole variety of considerations that go into a particular transaction, but the idea is to get the highest reward while managing potential deal-related risks.
First of all, thanks a lot for the explanation Numi. Such thorough discussions helps one to know about the kind of work performed at PE firms. As far as the explantion of kind of outsourcing work goes, you are very correct in your judgement and reasons. Almost, 75% of the work outsourced is about building models on the basis of historical data. The reasons provided by you are also very justified and practical. The analyst and associates based in the US have better idea about the markets there and are in a better position to take the best decission. However, in many cases (which are not exactly are the ones which involves buying and selling securities, rather than are deals which take time to settle), the client works with the support team in India to freeze the assumptions and decide the final results. This provides some exposure to the analysts based in the outsourcing location. And as far as the IP infringement issues are concerned, most of the outsourcing firms have NDAs with their clients and they also have a number of certifications for privacy issues. Also, many outsourcing firms also have IP division, so its highly critical for them follow such norms. Numi, I just wanted ask a question in general. I have got exposure to modeling and am quite efficient with it, however, for making a switch to a ER, PE or I bank, is it good to have certifications like what Wallstreetprep offers on the CV? I am a level II candidate (appeared for the same in June), so I am aware of the theoritical concepts associated with various models, however, don’t have hands on experience with many models. Though have experience of developing some of the models. Thanks again for the explanation and thanks in advance if you could spare some time and answer some of the queries I have.
It’s interesting to learn about the dynamic you guys have with the US-based analysts. Also, can you elaborate on what the IP concerns are? Do they relate to the financial models or something else, and how are they protected? For your question about getting a “modeling certification,” I don’t think that’s necessary. It’s much more important for you to be able to answer questions about models and show them that you know how to build one. I think if you can speak intelligently about modeling, that’d be fine. I’m not aware that anyone really cares about Bloomberg, Wall Street Prep, or whatever other certification there is out there – I sure don’t. The experience matters a lot more than your certification, so you should get as much hands-on modeling experience as you can. As a start, learn how to build a fully-integrated three-statement model from scratch.
numi Wrote: ------------------------------------------------------- > Good assumptions about a potential investment > drive the financial model, not the other way > around. agreed - important observation > That being said, the > model is still absolutely essential in order to > pinpoint valuation. it may be a matter of semantics - but ‘pinpoint valuation’ is sorta an oxymoron. intrinsic valuation by nature is a range concept, not a specific value. but yes, i think what you’re saying is a good model will allow you to identify a confidence interval more effectively.
Intrinsic valuation may be a “range concept,” since it’s as much an art as it is a science. That being said, when you’re submitting a bid for a company or a stock, you don’t say, “Well, I think I’d like to buy the asset in the range of $X-Y.” No – when you agree on a price, the amount you pay for something – anything – is a defined number. Maybe that’s in your valuation range, as it should be, but the number you’re paying is unambiguous, and that’s what I meant about “pinpointing” your valuation. There’s nothing self-contradictory about that.
Are there any outside programs/certifications I could look into for modeling? I saw “The Analyst Exchange” program on WallstreetOasis.com; I was curious as to some of your opinions on programs like that.
numi: That was really informative. Thanks.
bchadwick Wrote: ------------------------------------------------------- > Numi, that’s very enlightening. But I’m still a > little foggy on how “transactional” modeling > differs from “operational” modeling. Is it that > you’re making a decision about whether to do pure > equity, convertible debt, etc., and so you are > running a number of future scenarios to figure out > which has the best risk/reward? Or is it > something else. I realize I didn’t answer this particularly thoroughly, but since Ill be in the office till midnight and need to take a break anyway, I’ll tackle it now. Basically, the transactional model does like you said – we run a variety of scenarios to figure out which type of financing situation can give us the most reward while appropriately managing risk. As an example, let’s take a leveraged buyout (LBO) model. Assume that we already have the operating portion of the model built out (i.e. the part of it that tries to project the company’s operating performance in the form of P&L, balance sheet, and cash flow), we then want to evaluate how different financing considerations will affect IRR, the internal rate of return. Essentially, as a private equity investor, our objective is to get as high an IRR on our investment as possible, and laying on a transactional framework, namely the LBO structure, enables you to test how the IRR changes depending on your transactional or operating assumptions. There are several fundamental drivers of value to a private equity investor, namely (1) multiple expansion, (2) top-line growth and/or margin expansion, and (3) leverage. Our model reflects our assumptions for factors #1 and #2, and factor #3 is based on what we’re able to get from lenders as well as what the business can tolerate in terms of debt capacity. If you look at a private equity model in terms of these three aforementioned factors, it’s quite intuitive. Multiple expansion is probably the hardest to attain or even predict, but basically it refers to buying a company at some multiple of EBITDA and selling it in the future for a higher multiple thereof. You can do this by either getting a great deal on the company when you buy it, or through growth in the industry that makes the industry more attractive at the time you exit the investment. Even if EBITDA remained the same, an increase in multiple makes the exit valuation higher than the entry. Top-line growth and margin expansion are fairly self-explanatory as well – you want to have a business that’s expanding its earnings in some fashion so that even if the EV/EBITDA multiple is the same at exit as it is at entry (i.e. no multiple expansion), you can sell your investment at a greater price than what you paid for it. Finally, leverage, or the amount of debt, is your third and arguably the most critical element of your transaction. Obviously, the more debt you can use to finance the company, the better off you are as a PE investor – if you can service it and pay it down. But that is a big “if” – clearly, your equity value has more opportunity to expand if your initial equity injection is smaller, but you also need to carefully monitor how much debt you’re loading onto the business. And that’s where the financial model is so critical – it’s a representation of how much cash flow the company is generating, and helps you understand how constrained the business is going to be when you load debt on it. Plus, there are many different types of debt as well as management incentives that you have to take into consideration, and your model should be robust enough that you can test all these different transaction scenarios and see what your returns will look like. In a nutshell, these three key factors all help determine what your IRR is, and ultimately the private equity firm’s goal is to maximize IRR while being able to address its debt responsibilities. Thus, the whole purpose of your LBO model is to allow you look at how IRR changes as these variables change. Hope this helps…again, a long-winded response to how transaction models differ from operating models, but hopefully my “LBO Model 101” was useful.
Numi wrote: "Can you please go into more detail about the extent of your modeling responsibilities, and what you guys actually do as “outsourced modelers”? I have actually worked at not one, but two of the companies you mentioned above – and at least in my personal experiences, I felt that the modeling we outsourced overseas tended to be more mechanical in nature, i.e. setting up a model template, populating it with data from 10-Q and 10-K, and so forth. And while our counterparts overseas did a good job in this area, we (the analysts and associates in the US) typically retained responsibility for building the projections. " I do work on projections and key drivers. My PM goes through these and then asks me to make changes if he feels there is a need to do so.
Very enlightening post numi, greatly appreciated for those of us working on the sell-side ER departments of this world. Transaction work is definitely something I will look into a couple of years down the road.
btw, totally unrelated, what would be typical hours in PE? Similar / slightly less than IB (that’s my guesstimate)?
Numi, I’m curious if PE people use their model to help negotiate terms of their Credit agreements or are “market terms” just used. I’ve seen some companies come back to the table pretty quickly for amendments in order to do some basic liquidity transactions such as a sale-leaseback. Obviously these things can be overlooked or not anticipated. However, I know in HY analysis, CA agreement provisions are modeld in such as covenants or restricted payment baskets, etc. Are these thinsg considered when PE firms negotiate terms?
Numi, thanks a lot for the enlightening and knowledgeable posts. Its indeed helpful for us and will serve as a good tutorial for future reference. To answer your questions about IP protection and IP concerns, following is the explanation. IP concerns - These are concerned with both IP around financial models as well as IP around any other technology or intangible asset. For example, many I-Banks and Hedge Funds have their captive centers based in Outsourcing locations (mainly India, China, Latin America and Central and Eastern Europe) and they share their models with the support team, as far as my knowledge goes, many such teams help validate the models (for volatility and other major concerns) developed by the Mathematics gurus (PhDs and global mathematics scholars) sitting at the research centers of the banks and funds. Also, there are many technology or other patents shared with the support teams. As in my case, I am working with a technology venture group (they invest in various technologies and license it) of a major conglomerate firm based in US, and we help the client in finalizing the deals, so, for example, if the client is trying to acquire a specific portfolio of IPs of a company and wants to monetize the same by licensing it to other industry participants throughout the value chain (as they are a bigger firm and are in a better position to license the patents given their wide contact spectrum), at times they share the IPs with us and then we look at the market attractivess and the expected licensing revenues to be generated from the patents by licensing it out to a specific set of licensees in future. It involves going throug the market research reports, analyst reports and other secondary and primary sources to find the global market for that particular technology and then finding out the revenue on the basis of various assumptions and forecasts. It also sometimes involves generating the forecasted PnL, Balance Sheet and Cash Flow Statements for the new company formed (in case the client wants to have a major share in the company by aquiring the patents) and the expected revenue generated till the time (5 - 10 years) the client wants to move out. IP Protection - For protection, as I mentioned the companies have NPA with the clients and also there are various standards like ISO for data protection and confidentiality. I like the work I am doing now, but I want to move out of outsourcing, as the rewards and recognition here are very low. You can’t imagine the fees charged for the work done, its many times lesser than that charged in developed nations. The whole concept of outsourcing is based on cost arbitrage. I want to move to a good I Bank or a PE/VC firms but am not able to as my work experience is less (I have just completed a year) and I have a bachelors of technology in Aerospace Engineering, though I am a CFA, level 2 candidate (8th June 2008). Also, the major reason being that I lack contacts, the alumni base of my institute is strong, however, many of them are in big software firms in the US (many students from my insitute migrate to good universities in the US for MS and in the 90s there were many alumni who entered Silicon Valley). Some of them are in PE/VC firms and I banks, but then its a handful and they might not be having enough time to take into consideration the requests large pool of their juniors seeking their help. Again, thanks a lot for the insights about the PE firms and the work done there, and I shall be delighted to answer any further clarifications required.