Geoff Gannon Investor Questions Podcast #9: How Do You Calculate a Bank's Intrinsic Value?

Author's Avatar
Mar 18, 2010
Article's Main Image
The bank we’re going to look at is called Valley National (VLY, Financial). Valley National is a commercial bank with 198 branches in northern New Jersey and New York City. The bank is headquartered in Wayne, NJ. And the 198 branches are all within an hour’s drive of Wayne.

Over the weekend - Valley bought the deposits of two failed New York City banks from the FDIC. Because of that - Valley actually has more branches and more deposits than the numbers I’ll be using today. All the info I’m giving you here is out of date. The big difference is that Valley now has 600 to $700 million in added deposits. Along with those deposits - Valley picked up a couple extra branches in New York City. The 2 acquisitions over the weekend mean Valley is now about 7% bigger than it was in December.


Listen to the show

Let’s put the recent deals aside and just focus on Valley’s year end numbers. If what I say about Valley in this podcast gets you interested in the stock - you can always look into the deals yourself. For now - let’s stick to the numbers Valley gave investors in its annual report to shareholders and its 10-K report to the SEC. The first thing we need to talk about is why I picked Valley. In an earlier episode I read a quote from Warren Buffett . One of the things he said in that quote was “Don’t spend time on companies that don’t lend themselves to valuation.”

I think Valley lends itself to valuation. The reason why is simple. Valley is a conservative lender. And a slow grower. It looks pretty much the same from year to year and even from decade to decade.

A lot of banks are tough to value because they’re constantly changing. Maybe you don’t trust the management. Or the loan portfolio. Or the culture.

Finally, a big reason why I chose Valley is because I know Valley. I live in north Jersey. I know the towns Valley’s branches are in. And I know a more about Valley’s customers than I know about people living in Ohio or Illinois or Nevada.

That’s where you start with a bank. You look for the familiar. You look for a place you know. You look for a management team you know. Or for a culture you’ve watched and haven’t seen change.

That’s where you start when you’re looking for a bank stock. You start with the intangibles. The soft stuff.

Most of this episode is going to be about the hard stuff. We’re going to talk facts and figures. We’ll spend a lot of time with the balance sheet and the income statement.

But that’s never step one when you’re looking for a bank stock to buy. I’ve read past annual reports from Valley. I’ve read - and watched - interviews with the CEO. I have a better feel for Valley than I do for most banks.

And that’s where you start. You start by picking a bank that lends itself to valuation. You don’t start with something that’s got too many complications, too many moving parts, or too much change.

You start with something simple.

Okay. Enough with the intangibles. Let’s talk numbers.

Whenever I look at a bank - I start with the deposits. On page 74 of Valley’s 10-K you’ll find the company’s balance sheet for December 31st, 2009. Look under liabilities and you’ll see Valley ended last year with $9.547 billion in deposits.

Deposits are listed as a liability - but they’re really a bank’s most valuable asset. Just like float is listed as a liability on an insurance company’s balance sheet - deposits are listed as a liability on a bank’s balance sheet. But unlike float - deposits always cost banks money. I don’t know any bank that makes money on deposits before lending them out.

That’s different from an insurance company’s float. Some insurance companies - like Progressive (PGR, Financial) - have combined ratios under 100. That means they earn an underwriting profit and still get to invest the float.

That doesn’t happen at banks. All banks lose money gathering deposits. Banks don’t make money when you give them your cash. That part of the business loses money for them. Banks provide more services than they charge you for. They make up the difference - and earn a profit - by lending your money out at rates high enough to offset the costs of taking in your deposit.

At year end Valley National had more deposits than gross loans. Deposits were $9.55 billion and gross loans were $9.40 billion. Not all banks have more deposits than loans. The year before - Valley had lent about $800 million more than it had in deposits. That’s normal for banks that use other sources of borrowing.

But deposits are the best funding source. Banks don’t make a lot of money borrowing and lending to others - because they have to borrow the money at higher rates than they pay on customer deposits.

This will be obvious in a minute when we go through Valley’s costs. I just want to remind you that banks do have other sources of cash than deposits. But those sources are not good money makers. And they do not create goodwill. When we talk about goodwill we’re really talking about what a bank is worth over book value. A great deposit gathering bank can be worth a couple times book. Just like a great underwriting insurance company can be worth a couple times book. But an insurance company that writes at a combined ratio of 105 is not worth more than book. The same is true of a bank that borrows money and re-lends that money instead of using customer deposits. That bank is doing the same thing as an insurer with a combined ratio of 105. It’s not value.

There are only 3 ways a bank can make by lending. The first is to gather deposits at a cost well below the market price of money. This is what most banks do. They provide services to their customers that draw cash into the branches. The customers put the money there not because of the interest they earn but because of all the great service they get. It’s convenient. It’s useful. And if the bank provides great service at a low cost - it can basically buy money from its customers at less than it would cost to buy money from anyone else.

As I’ll show you in a minute - Valley bought money from its customers at 3.39% last year. It then lent that money out at 5.74%. Even triple A corporate borrowers pay around 5.24% right now for long-term money. Technically, deposits are short-term money. They can be pulled at any time. But in reality - a bank’s customer deposits are a lot like an insurance company’s float. As long as the bank is growing - total deposits are basically one giant permanent loan.

If a bank can gather deposits as cheaply as Valley did in 2009 - it can make about 1.85% a year just by lending at triple A corporate rates. Even a bank that only did the safest mortgage lending would still make more than 1.6% a year on its deposits - if it gathered those deposits as cheaply as Valley did.

And that’s the key. A bank has to gather deposits cheaply. If it doesn’t do that - the only way to make money is to take big risks.

Banks that gather deposits as cheaply as Valley can take almost no risk on the lending side and still make 1.5% to 2% a year on their deposits. That’s because Valley gathered deposits at a cost of 3.4% last year - and even the best long-term borrowers paid 5% a year and up in 2009.

All of this is just background to explain why we’re going to calculate Valley’s intrinsic value based on its deposits.

Deposits are what makes a bank money. Everything else is either temporary or unlikely to add much value over book. So when analyzing a bank stock - start with deposits to try to find that bank’s economic goodwill. If you don’t find any - chances are the bank isn’t worth more than book value.

Valley is worth more than book. And here’s how we prove it.

We start with the $66.09 per share in deposits. That’s how much Valley had at the end of 2009. So when you buy a share of Valley’s stock - what you’re really buying is $66 in deposits.

The next question to ask is: “How much do those deposits cost?”

Deposits are only valuable when they cost the bank less than the rates they can be lent out at. Right now I don’t want to get into what rates a bank can lend at. So let’s just use the triple A corporate bond yield. Triple A corporate bonds yield 5.24% right now.

So for deposits to be clearly valuable they need to cost less than 5.24%. I should point out that while the triple A corporate rate sounds like an odd choice - it works fine in this case. It’s between the lowest rate on mortgage lending - about 5% - and the actual rate Valley lends at which is 5.74%.

When you first look at a bank - you don’t know how risky its loans are. And you don’t know how big the losses will be. The same thing is true of mortgage lending generally. I don’t have an opinion about house prices. Do you?

If not - it’s best to use the triple A corporate bond yield. That’s as close as we can get to a non-government risk free long-term interest rate.

The reason we don’t use government bonds is that buyers of government bonds often have different reasons for buying and selling than we do. Corporate bond buyers are more like us than Treasury bond buyers. So I like to use the triple A corporate bond yield as a yardstick for the price of long-term money.

It’s a good rate to use when looking at Valley. Remember this is a commercial bank we’re talking about. Valley makes all kinds of loans. Everything from commercial mortgage loans to airplane loans.

Now that we have one side of the value equation - the interest rate Valley can lend at - we need to look at the other side. The other side of the equation is what it costs Valley to gather deposits.

From now on I’ll be working off pages 74 and 75 of Valley’s latest 10-K report to the SEC. You can find this report at the EDGAR website. Page 74 is Valley’s balance sheet. And page 75 is Valley’s income statement. We’re going to move back and forth between the two statements - using numbers from one and dividing them by numbers from the other.

If you’re used to valuing non-financial businesses by their cash flows - this will be a new experience for you. It’s simple. It just sounds strange at first.

Try to follow along. And pay attention to why I’m doing what I’m doing. Don’t worry about the exact numbers. Just try to follow the steps. And try to understand why you should look at a bank this way.

The first thing you do is get out a pen and a calculator. You don’t need a separate piece of paper - just write in the margins of the 10-K. There’s always lots of whitespace in SEC reports. I scribble all over them.

The first numbers we’re going to look at are average deposits and average loans. For loans I’m going to use the gross number. In other words - for 2009 - I’m going to take the $9.268 billion in net loans and add back the allowance for loan losses of $101.99 million. In this case - you don’t even have to do that - because Valley lists “loans” separate from “net loans”.

Ignore the net loans number and use the “loans” number instead. It’s $9.37 billion for 2009 and $10.14 billion for 2008. Since we’re going to compare Valley’s loans to the interest earned on those loans in 2009 - we want to take an average of the starting and ending amounts. That means you add $9.37 billion and $10.14 billion and then divide by 2. You get $9.76 billion. That’s Valley’s average gross loans during 2009.

Now go to the income statement. The very first item - under Interest Income - is called “Interest and fees on loans”. That’s the number we want. It’s $561.25 million.

Take that $561.25 million in loan interest and divide it by the average gross loans of $9.76 billion. You get 5.74%. That may be a bit lower than the actual rate Valley is lending at because we used gross loans instead of net loans. Don’t worry about it. The difference is small. And we’re using the more conservative approach. When in doubt - always use the most conservative approach. And when it comes to analyzing a bank - you’re going to be in doubt most of the time.

Okay. So write 5.74% down somewhere. That’s what Valley can earn on its loans. At this point we might also want to check total interest divided by investment securities plus loans. I’m not going to make you do that. Because in this case the number is close. Securities might pay more like 5.4% versus 5.74% on loans. But it’s clear we’re in the right ballpark. So let’s just leave investment securities aside and focus on loans and deposits.

Again - Valley is lending at 5.74%. Now we want to know how much it costs Valley to gather deposits.

This is the most important part of what we’re doing today. Most of a bank’s value comes from its cost of deposits.

Start by looking at the balance sheet on page 74 of Valley’s 10-K. Under liabilities - you’ll see total deposits of $9.55 billion in 2009 and $9.23 billion in 2008. We want the average. So add them together and divide by 2. You’ll get $9.39 billion. That’s Valley’s average deposits for 2009.

Write that number down somewhere. Then go over to the income statement. Go to interest expense and you’ll see two lines listed for “interest on deposits”. Add them both together. You get $118.3 million.

Divide that by Valley’s $9.39 billion in average deposits. You get 1.26%. That’s how much interest Valley paid on its deposits in 2009.

That number isn’t very interesting. A lot of banks pay roughly the same interest rate on customer deposits. What matters most is the bank’s non-interest income and non-interest expenses.

Non-interest expenses are bigger than non-interest income for most banks. So we’ll start with non-interest expenses and then subtract non-interest income.

There are 2 ways to do this. One is the easy way. The other is the right way. The right way is to back out all the things that have to do with trading instead of deposit gathering. Also you want to reverse any amortization of intangibles. Amortization is a non-cash charge without much point in the real world. And finally - in this case - I would also use the 2008 FDIC assessment instead of the 2009 FDIC assessment. Last year included a special charge from the FDIC that was much bigger than it will be in the future.

Don’t worry I’m not going to make you do all that - I’ll just tell you what you get. Valley’s 2009 non-interest expense goes from $306.028 million down to $280.998 million. And non-interest income goes up from $72.251 million to $81.019 million.

In other years we would’ve gotten different results. Sometimes the adjustments would give us lower numbers. Here most of them are in Valley’s favor.

So now we take the adjusted non-interest expense of $280.998 million and subtract the adjusted non-interest income of $81.019 million. That gives us $199.98 million in adjusted net non-interest expense.

That’s a mouthful. But it’s an important number. What we’re doing here is calculating how much Valley spends on things other than interest to gather customer deposits. To get the answer we just divide the adjusted net non-interest expense of $199.98 million by Valley’s average 2009 deposits of $9.39 billion. When you do that - you get 2.13%.

That’s how much it costs Valley - in addition to what the bank pays in interest - to keep customer deposits.

Next we want to add the two different costs together. Valley’s interest cost on deposits is 1.26%. And its non-interest cost on deposits is 2.13%. That means Valley’s total cost on deposits is 3.39%.

You can think of that 3.39% as the rate Valley borrows at. Or - if you prefer - you can think of its as Valley’s cost of production. Valley can mine money at a cost of 3.39 cents per dollar.

The market price for money is 5.24 cents per dollar. That’s if we use the yield on triple A corporate bonds. I think that’s a good starting point.

Now let’s calculate Valley’s pre-tax deposit earning power. Take the 5.24% yield on triple A bonds and subtract the 3.39% it costs Valley to gather deposits. You get 1.85%. That’s Valley’s economic value margin on deposits.

Simply put: Valley mines money at 3.39 cents per dollar and then sells that money in the open market at 5.24 cents per dollar.

That’s just a metaphor. As you know - Valley makes the loans money itself. And it makes them at 5.74% - not 5.24%.

But for now we’re going to do a pure cost of production calculation. So we’ll use the market price of money as measured by triple A corporate bond yields.

Once again: that means 5.24% minus 3.39% equals Valley’s economic value margin on deposits of 1.85%.

To calculate the intrinsic value of Valley’s stock we start with Valley’s deposits per share. You can find this yourself by going to the balance sheet and taking total deposits and then flipping over to the income statement to find diluted shares outstanding. I actually did this for you earlier. The answer is $66.09 in deposits per share.

Now we have a way of valuing the deposits. We can figure out how much the deposits per share are worth by multiplying the $66.09 a share in deposits times Valley’s economic value margin of 1.85%. Remember - that 1.85% is how far below triple A corporate bonds Valley’s cost of production is. Valley produces deposits at 3.39%. And it should be able to lend them at 5.24% or higher.

The value of a bank’s deposits comes from this spread between the market price of money - which is basically a commodity - and the bank’s cost of production. The difference in this case is 1.85% times the $66.09 a share in deposits. That comes out to $1.22 a share in pre-tax deposit earnings power.

Like we did with insurance companies - we’re going to multiply that pre-tax number by 10. Remember multiplying pre-tax earnings by 10 is a lot like multiplying after-tax earnings by 15. So we’re kind of putting a P/E ratio of 15 on these deposits.

That $1.22 a share in pre-tax deposit earnings power times 10 gives us an intrinsic value of $12.22 a share.

That’s a conservative estimate of what Valley’s stock is worth. It only looks at deposits. And it uses the triple A corporate bond yield instead of the 5.74% Valley actually charges on its loans.

A more realistic intrinsic value estimate would be $15.53 a share. That comes from assuming Valley will lend its deposits at 5.74% instead of 5.24%. Of course there might be losses.

Valley has a great record when it comes to lending. You can read about it in the company’s annual report. Any analyst who talks about the company will tell you the same thing. Valley’s lending standards are tough. And they’ve been tough for decades.

So it might be okay to assume Valley will lend at 5.74% and have very low loan losses in the long-run.

But the best way to look at any bank is to start by calculating the bank’s cost of production and then seeing how far below the triple A corporate bond yield that cost of production is.

You then multiply that number by the amount of deposits per share. And then multiply that pre-tax earnings number by 10.

If you can buy a bank like Valley at less than that conservative intrinsic value estimate - you should consider it. When I say a bank like Valley - I mean a bank with a long history of conservative lending.

That makes it easier to value the bank. You can focus on deposits instead of worrying about loans.

And deposits are where the value is at any bank.

So - to sum up - we did an intrinsic value calculation for Valley based only on deposits and decided the bank was worth between $12 and $16 a share.

Realistically, it looks like Valley is worth about $15.50 a share based on its deposits alone. And conservatively I felt Valley was worth $12.20 a share at the bare minimum.

Just like when you look at any other business - there can be other nuggets of value in a bank. For example: Valley owns about 100 plots of land - and the buildings on them - throughout northern New Jersey. It looks like Valley carries the land on its books at a bit under $600,000 per plot.

Is the land worth that much? Or is it worth a whole lot more?

Personally I think it’s a mistake to get hung up on that sort of thing. Valley uses the land for its branches and - in 4 cases - for offices. The value in a bank like Valley isn’t in the real estate it owns. It’s in the customer deposits.

That’s what you focus on. You think about customers. And you think about deposits.

Almost all of a bank’s value - and literally all of a bank’s value over book - comes from customer deposits.

So that’s always where you start when you calculate a bank’s intrinsic value. With deposits.

Okay. That’s all for today’s show. If you have an investing question you want answered call 1-800-604-1929 and leave us a voice mail. That’s 1-800-604-1929.

Thanks for listening.

5 / 5 (28 votes)
Load More
Author's Avatar

GuruFocus Screeners

Related Articles