Why Would You Not Use a DCF for Financial Institutions?
The guide says it's because fin institutions are highly levered and they do not re-invest debt in the business and instead use it to create products. Also, interest is a critical part of a bank's business model and working capital takes up a large part of their balance sheet.
What exactly does this mean? Is there a better answer to this?
And what are some good reasons why you would not use a DCF? Other than having unpredictable/unstable cash flow and for financial institutions.
Thanks.
Valuing Financial Institutions and Banks
From the Financial Institutions Group or FIG is an industry group that focuses on providing advisory services to financial institutions.
Understanding why financial institutions groups are different from traditional coverage groups is key. This is because the companies are different from traditional cash flow driven companies.
The financial institutions group offers advisory to many diffrent financial firms
These firms include:
- Banks
- Insurance Companies
- Financial Technology Companies
- Asset Managers
- Diversified Financial Companies
- Securities Firms
- Other Specialty Finance Firms
In this way the financial institutions groups are unique because of the unique asset classes involved, similar to energy and real estate.
Specifically, FIG client assets can include:
- loans
- investments
- cash
- securities
Basically, FIG companies "borrow money" (i.e., source capital) cheaply and then "lend money" (invest that capital) expensively. Much (though not all) of their income is generated by the spread between those two rates of return.
Additionally, because of the sources of some of their capital (largely individual consumers), there is strict regulation surrounding what kind of assets FIG companies can and cannot hold on their balance sheet, and in what quantities.
valuation using dividend discount model
DDM - Dividend Discount ModelTraditional banks have to maintain a certain level of liquid assets, which means that at any given time, they have to be above threshold levels for several capital ratios (i.e. in case depositors withdraw all deposits, liquidity sources disappear, etc). Thus, traditional net income (and therefore FCF) is a pretty bad place to start.
- You should start with net interest income, which is essentially the bank's gross margin (think interest income as revs, interest expense as COGS... this should be intuitive).
- Next, look at a bank's non-interest expense (fixed costs, rent, and the biggest part--comp expense). You also have to keep in mind that the bank has to keep making loans to make money (its "assets" are earning), so there are certain ratios that your model should be maintaining (asset/loan ratios, assets/equity, etc.).
- Now, after mandatory debt repayment, reserves, and all the jazz, you should get to a net-income-like figure, but more accurately, it's "excess capital available for distribution." That's the "dividend" that gets paid out in the DDM / Gordon Model.
- Like every PV method, it works the same way as a DCF. Of course, there are other quick-and-dirty methods you can apply; comps, price-to-book-value ratios, deposit premiums, etc.
I glossed over some of the fuzzier aspects, but this should be a reasonably accurate overview. Recently, I think there were some pretty laughable rumors of a Bank of America sale, but just for a second--think about the complexity of that model. Take your traditional banking model, latch on an investment banking business, mortgages, credit cards, etc... It would SUCK to be the analyst modeling that one. The complexity of this model is also the reason that people say FIG is more sophisticated than other industry groups.
Recommended Reading
Preparing for Investment Banking Interviews?
The WSO investment banking interview course is designed by countless professionals with real world experience, tailored to people aspiring to break into the industry. This guide will help you learn how to answer these questions and many, many more.
Well, the answer I gave when I interviewed at Sandler was that "most banks nowadays dont even have positive earnings/cashflow/bottomline/anything, so what good does a DCF do when your NPV is less than 0?"
Keep in mind that commercial banking isnt doing so hot, and that is their bread and butter.
I could be wrong, as I didnt get that job, lol. But thats my take.
I hate financials, it's like the twilight zone of valuation. Truth is, no one knows how the hell to value them anymore, especially in this market. Flip a coin, if heads, buy, if tails, sell.
It basically has to do with what you stated. They both borrow and lend, so you can't treat debt and interest expense the same way as you would for a non-financial entity. In some sense, interest expense is more like COGS, they're looking to make money off interest rate spreads, that's like their "gross margin". Because they're so opaque and it's very hard to figure them out, people dumb it down and go by the book value instead.
Think through it practically:
CF how would you split between operating result and financial result for a bank? You can't. So how do you compute FCF? You can't.
WACC: how do you compute cost of debt? They will have at least 100 different types of debt including 50 types of savings accounts and there is no way you can reasonably forecast how they would react to a change in target capital structure.
Target capital structure: For banks, equity ratios are based on risk weighted assets rather than assets. So how do you set a target d/e ratio? You can't.
Similar goes for insurance companies.
If you're REALLY interested and want to through down $100 and a few days of your life, then check out the breaking into wall street financial institutions modeling tutorial... it explains a lot
Solidarity,
Traditional banks have to maintain a certain level of liquid assets, which means that at any given time, they have to be above threshold levels for several capital ratios (i.e. in case depositors withdraw all deposits, liquidity sources disappear, etc). Thus, traditional net income (and therefore FCF) is a pretty bad place to start.
I understand most part you were saying except why FCF is a pretty bad place to start even if they have to maintain some threshold levels? Can you elaborate a little bit more? Much appreciated!
Think the way to do a dividend discount model is to figure out a capital threshold and then any "excess" over that is the dividend that you discount back to the present to come up with an equity value.
Quote Solidarity. DDM with Excess Capital. Talk about this one during an interview with a JPM FIG banker and he agreed too.
FIG valuation - betas (Originally Posted: 12/03/2015)
Hi guys, I am doing some valuation work for a private asset manager. I need to do a simple DCF based on their business plan and for this I would need the Cost of Equity.
In order to do this, what would make most sense? Would I gather a list of listed comp companies, calculate their (levered) beta, take the average beta of my comp set, unlever it and then relever it to the cap structure of the firm? But how does unlevering and relevering work in the case of financial instutions, where the D/E ratio plays a different role vs. traditional firms?
There's a two-part answer to this:
You don't value financial companies using a DCF. I'm not too familiar with AM, but for most other FIG companies, such as banks or specialty finance companies, a DCF serves no purpose as cash flow is not a metric used to value them. Hopefully someone with more knowledge in the FIG space can expand on this.
The ideal way to calculate beta would be to compile a list of comps (I normally use five to ten, but my group has an ongoing list, so it's not like I have to hunt them down every time) and then calculate the betas. Unlever those betas, average them (calculate the median for comparable purposes too), and then relever that average using the target's capital structure. Keep in mind that while a DCF is meant to get to a valuation, it's usually used to justify one, meaning that an MD might say "I think Company X is worth somewhere between $Y and $Z", and then it is the job of the analyst to make reasonably assumptions to get the DCF to spit out a range between $Y and $Z.
Thanks. Yes you are right, instead of a DCF you would aim to do a DDM. But does relevering betas for financial institutions work the exact same way as for companies?
With banks and insurance companies, comparable companies are similarly leveraged, so there's no need to un/relever betas.
My understanding is that asset managers are actually not too dissimilar from normal companies and can be evaluated via normal DCF/EBITDA/etc. methods you can use for a widget company.
It's the same in banking, specialty finance, really any FiServ company. E.g. if one bank has a lev ratio of 2.5 and one is 3, that's going to be factored into their individual betas. Honestly though I don't think you really need to break out CAPM. You could just find an average CoE for small public AMs and add 2-3% as a size/liquidity discount.
Also, I assume you're doing comps, but industry-specific multiples like EV/AUM would be pretty helpful as well.
Why would you use P/BV for a financial institution? (Originally Posted: 12/12/2015)
Why would you use P/BV for a financial institution?
For financial institutions, a lot of their assets and liabilities are financial instruments that are marked to market, so book value should = market value there. Loans for a bank aren't marked to market, but are supposed to be adjusted for risk of default via a loan loss reserve.
In short, if price is a lot lower than BV for a FI, then the market either doesn't trust that the assets have been properly marked to market, or there are hidden liabilities that have been pushed off the BS. If market is much more than BV, market is bullish on growth, brand, etc. Today a lot of banks trade around 1.5x BV.
Thanks! What do you mean by "marked to market"?
Some banks have extremely high ( 3+) or low ( below 1) . Does that imply their being over or under valued ?
Similique rerum harum iusto voluptatem ut et. Consequatur aut velit quia fuga. Repellat in voluptas doloremque et.
Placeat voluptatem quia iste itaque. Dolor corrupti excepturi et qui in ea qui et. Quia ut et velit non. A ad rem et maxime.
See All Comments - 100% Free
WSO depends on everyone being able to pitch in when they know something. Unlock with your email and get bonus: 6 financial modeling lessons free ($199 value)
or Unlock with your social account...