GuruFocus Premium Membership

Serving Intelligent Investors since 2004. Only 96 cents a day.

Free Trial

Free 7-day Trial
All Articles and Columns »

Geoff Gannon Investor Questions Podcast #13: How Do You Find a Stock's Z-Score?

June 11, 2010 | About:
Geoff Gannon

Geoff Gannon

411 followers
Yesterday I talked about the F-Score. Today I’m going to talk about the Z-Score. The Z-Score was created 42 years ago by a New York University Professor.

He wrote a paper on using financial ratios to predict corporate bankruptcies. The Z-Score is simple. In some ways it’s simpler than the F-Score. In other ways it’s not.

The Z-Score looks simple. Because it only has 5 variables. The F-Score has 9. But the F-Score is a checklist. The Z-Score is not. All F-Score questions are worth 1 point.

Z-Score questions have weights. That means they are worth different points.

The F-Score was created without any statistical analysis. It just uses common sense. The Z-Score does not use common sense. It uses statistics. The professor checked a bunch of different variables against a database of bankruptcies and used the ones that made the best predictions.

The F-Score was created to pick stocks. The Z-Score was created to predict bankruptcies. So there’s no promise a higher Z-Score means a better stock. It just means the stock is less likely to go bankrupt.

The Z-Score is made up of 5 questions. The first question is working capital divided by total assets. Working capital means net current assets. Go to the balance sheet. Find current assets. Then subtract current liabilities. That’s working capital. Divide it by total assets. And that’s your answer to question number 1.

That brings up another difference between the F-Score and the Z-Score. The F-Score had 9 questions with answers of 0 or 1. The Z-Score has 5 questions with answers that can be anything. It’s not a checklist. It’s an equation.

To figure out a stock’s Z-Score you should use Microsoft Excel. If you don’t know how to use Excel it’s easy to learn. But if you don’t want to learn you’ll need an index card. Write down the equation for the Z-Score. List the 5 questions and their weights. I’ll give the weights at the end of the podcast.

But don’t try to listen to this podcast and then do just one Z-Score. Practice on a bunch of different companies. Practice as much as you can. Anytime you’re interested in a stock figure out its Z-Score. It’ll get easier and easier for you. Pretty soon you’ll be able to figure out a Z-Score in a couple minutes. Just write down the formula somewhere. You don’t want to keep going back to check if your weights are right.

Let’s stop here and talk a bit about why the Z-Score includes working capital divided by total assets. The easy answer is that it got the bankruptcies in the database right more often than it got them wrong. At least when used with other numbers. In other words the statistics worked for the data the professor was using.

That’s true. And that’s the real reason the number is included. But a doctor wouldn’t stop there. Statistics show taking Advil lowers your risk of getting Alzheimer’s. But those are just statistics. And a doctor wouldn’t go right out and start telling his patients to take as much Advil as possible.

We want to know if there are other differences between people who take a lot of Advil and people who take a little. We want to know why people take Advil and why they don’t. We want to know what the risks of taking Advil are. And what are the risks of taking a lot of Advil. Also – how does Advil lower your risk of Alzheimer’s. Correlation is not enough. We want to know the cause.

But do we need to know the cause? The honest answer in investing as in medicine is usually no. We’d like to know the cause. But if our job is reducing the risk of Alzheimer’s or reducing the risk of bankruptcy and we’ve got nothing better to offer than Advil or a lot of working capital – well then we go with what little we know. And if we know it’s not too risky to take a small amount of Advil then we tell the patient to do that. And if we know it’s not too risky to buy stocks with a little extra working capital then we do that too.

My point is that knowing the cause of something is helpful. But knowing there’s a correlation is better than nothing. And if our options are something or nothing then we go with something. We go with a correlation.

The working capital to total assets issue is not that well understood. Working capital are things like cash, receivables, and inventory. Some great companies that have no risk of going bankrupt don’t have much working capital. And some bad businesses do have a lot of working capital.

The theory is simple. Think back to my bathtub metaphor from yesterday. I gave you the standard lecture on stocks and flows. The water in the tub is our stock of cash. And the flow is the amount of water pouring from the faucet minus the amount of water going out the drain. The goal is to keep the tub from running dry. If the tub runs dry you go bankrupt.

That’s the theory. It doesn’t hold up perfectly in the real world. You can look up a lot of papers written by accounting professors on whether cash is most important or current assets are most important. Different papers give different results. I’d say the truth is murky. It’s probably different for different industries. It’s definitely different for different companies.

One reason is that companies are run by human beings. They don’t behave randomly. They have a purpose. They have goals in mind. And they have fears. Like a fear of bankruptcy. Some companies hoard cash because they know their business is shaky. Some hoard cash because they know their industry is headed for hard times. Others hold cash because they know they’re in a buggy whip business. They know the future will never be as good as the past.

Fine. So sometimes managers have good reasons for holding cash. But just as often they hold cash because they lived through the Great Depression or Japan’s lost decade or some other horror story. And know they’re free from the horror but it’s all they think about at night.

If that’s true – if managers are being paranoid – well then holding cash is a sign of strength. No one doubts that more cash is better than less cash. But sometimes there is a reason for a company holding cash. So more cash may not be as good a sign as it seems to be.

That’s why we use statistics. It’s why we check the data. We don’t just think about theory. We actually dig into the history.

And what does history tell us? It says that when combined with these other variables companies that have higher working capital as a percentage of total assets are less likely to go bankrupt. Simply put more working capital is better than less working capital.

But that’s like saying more Advil is better than less Advil. It’s just a statistic. It’s not a promise. And you can’t rely on that one thing alone. You have to look at all the different risks.

So let’s move on to Question #2. Question 2 is a great example of causes versus correlations. It’s one of the best variables in the bunch. But to a lot of people it doesn’t make a bit of sense.

Question 2 is retained earnings divided by total assets.

First let’s talk about retained earnings? What are they? Retained earnings are earnings not paid out in dividends. Losses are also retained. So really “retained earnings” should be called “net retained earnings”.

Why do retained earnings matter? They’re the most backward looking number you’ll ever find. Bankruptcy is about the future. Not the past. So why do past earnings matter?

Warren Buffett’s teacher, Benjamin Graham, was big on looking at the past to see the future. I am too. I’m sure you’ve heard that past results are no guarantee of future returns. That’s true. But history is important.

A history of losses is important. A history of profits is important. A history of risky behavior is important. So is a history of dumb acquisitions. All these things matter when we look at stocks, industries, or CEOs. They even matter when you look at countries. One of the best guides to whether a country will default on its debt in the future is its history of default. Right now we’re worried about countries like Greece and Portugal. Guess what? Greece and Portugal have bad credit histories.

In the 1800s: Portugal defaulted in 1828, 1837, 1841, 1845, 1852, and 1890. Greece defaulted in 1826, 1843, 1860, and 1893. Since the 1800s, Portugal has been in default 10% of the time. Greece has been in default 50% of the time.

Greece and Portugal’s bad credit histories help explain why bond markets don’t trust them to have debt-to-GDP ratios as high as the United States, United Kingdom, or Scandinavia.

Yes it’s unfair. But it’s also logical. You don’t trust people to pay you back when they’ve proven time and again they won’t. And you do trust people who prove time and again they’ll do what it takes to pay you back.

Maybe this time is different. I won’t argue it’s not. But if two countries or two companies have equally bad balance sheets and one has a history of paying its debts while the other doesn’t, I’ll go with the good credit history every time.

Earnings work the same way. Some people thought I was too hard on Evergreen Energy (EEE) when I said it was going to go bankrupt. Maybe I was too hard. But I don’t go soft on companies with long records of losses. Evergreen isn’t just in a bad place right now. It has a long record of losses.

Even a company with a long history of profits can run into problems. A few have even gone bankrupt. Why should we believe a company with no history of profits will have a deathbed conversion. A bad company is unlikely to become a good company just because it needs to pay the bills.

It takes a lot to turn a company around. Bankruptcy doesn’t mean the management was bad. It doesn’t mean they wanted to line their own pockets. It just means they failed. And when the Z-Score shows you a number like 0.6 it’s telling you that company is going to fail.

You need to look at Z-Scores that way. Take your emotions out of it. All the evidence is against Evergreen Energy. The only thing on the other side of the scales is hope. And hope isn’t an argument.

You need to be honest with yourself when you invest. That’s the biggest challenge for any investor. And the Z-Score helps you do that. But if you doubt it in as clear a case as Evergreen Energy you need to rethink the way you look at stocks.

Ben Graham said you’re neither right nor wrong because the market agrees with you. You’re right or wrong because your reasoning and analysis are correct. It’s no use being contrary just to be contrary. That’s no better than just going with the herd. You need to be independent. And you need to be honest.

The Z-Score can help you do that. But it can only help you if you’re willing to look at the cold hard facts. It can only help you if you’re willing to take emotion and hope out of the equation. When I talked about Evergreen Energy I did that.

And yes that meant I crushed that poor caller’s hopes. But I did it because the facts told me to do it. I did it because he still owned a stock I thought was worthless. And it was my job to do for him what he needs to learn to do for himself – what every investor needs to learn to do – be honest.

And the honest answer is that Evergreen Energy will go bankrupt. I know the management hopes it won’t. I know they’re trying. But the managements of all bankrupt companies tried. And the managements of all bankrupt companies had plans. They failed. And Evergreen Energy will fail too.

I don’t want to beat up on Evergreen Energy. I was asked a question by a listener. And I want to be honest with listeners.

The Z-Score can help you be honest with yourself. It can help you replace hope with facts. The F-Score can too.

But in many ways the Z-Score is more important. Because the denial I’ve seen from investors in companies about to go bankrupt is scary. It’s beyond what I’ve seen in troubled companies.

A lot of times investors know the company they’re investing in is troubled. They know it has problems. They see the good points and the bad.

But they never admit it’s going to go bankrupt. They always hold out hope that the things management is saying will come true. Maybe the CEO believes it. But in a way he’s paid to believe it. He’s paid to try. You’re not. He has to go down with the ship – reassuring the crew all the way. You can get off the boat.

And you should. Even though it means taking a big loss. You should. And that’s why the Z-Score is so important. It’s a reality check. A dose of sanity when you need it most.

The reaction to my podcast on Evergreen Energy reminded me how important the Z-Score is. Because I almost forgot how far in denial folks can get. They think future plans make up for a history of losses. They think the chance of unlocking some value makes up for the certainty of debts coming due.

It doesn’t. Not because the CEO is a bad guy. Not because the employees aren’t doing everything they can. It doesn’t because that’s the way the world works. Unless dreams become reality a bright future doesn’t make up for a dismal past.

The Z-Score puts that into numbers. And if it’s easier for investors to face facts when those facts are put in numbers than the Z-Score can work wonders for how honest you are with yourself. And it can keep you from losing everything.

Okay. Enough with the rant. Back to Question #3. Question 3 is earnings before interest and taxes divided by total assets. Earnings before interest and taxes is sometimes called EBIT. It’s a measure of earnings that can be used to pay interest on the debt. So it’s a lot like using a company’s net income. Only better. Because net income is the amount left over to pay shareholders. This is the amount left over to pay debtholders.

Some people like using EBIT divided by enterprise value to find a stock’s earnings yield. I don’t always agree with that. But I do agree that EBIT is the number to use when we’re talking about bankruptcy risk.

Question #4 is the market value of equity divided by book value of debt. That means you take the number of shares outstanding – you can find this on the cover sheet of the stock’s 10-Q or 10-K – and you multiply it by the share price. That’s the market value of the equity.

The book value of the debt is the same thing we talked about with the F-Score. I would take “long-term debt” plus the “current portion of long-term debt” plus anything called “short-term debt” or really anything with the word debt in it. When in doubt count it as debt. But you shouldn’t be in doubt all that much. Current liabilities are not debt. Deferred taxes are not debt. You get the idea.

The last question is an odd one. Question #5 is sales divided by total assets. You might remember this from the F-Score. It’s called the asset turnover ratio. It was a popular ratio in Ben Graham’s day. It’s less popular now. Some of that may be because it varies a lot from industry to industry. For example: railroads have low asset turnover ratios. An asset turnover ratio of 0.5 is not bad for a railroad. It’s terrible for a retailer.

That might be why you never hear about the asset turnover ratio. However, the asset turnover ratio can help predict bankruptcies when combined with other variables. That’s why it’s included in the Z-Score.

Now for the weights. I’ll put a link in the “notes” section of the website. If you follow that link you can see the weights for each variable.

The weight for working capital divided by total assets is .012. The weight for retained earnings divided by total assets is .014. The weight for earnings before interest and taxes divided by total assets is .0333. The weight for market value of equity divided by book value of debt is .006. And the weight for sales divided by total assets is .999.

To get good at using the Z-Score you should create a worksheet in Microsoft Excel that does it automatically for you once you paste the right numbers in. You’ll still need to go to the 10-Qs and 10-Ks yourself. But the ratios and their weights will be calculated for you. If you don’t know how to use Microsoft Excel you should write down the weights on an index card and keep it on your desk. You don’t want to keep searching for the variable weights. You want to get straight to actually figuring the Z-Score for stocks you’re interested in.

To help you with that I’ve calculated the Z-Scores for a couple stocks we mentioned. Those stocks are Omnicom (OMC) and Birner Dental Management (BDMS).

If you’re listening to this podcast in the future – that is, not on Friday June 11th – these Z-Scores won’t match exactly. That’s because the stock price is part of the Z-Score. So as the stock price changes, so will the Z-Score.

But you should be able to get a Z-Score in the right ballpark. So once you’re done listening to this podcast, I suggest you try to calculate the Z-Score for each of these stocks and see if it matches the numbers I’m about to give you.

Omnicom’s Z-Score is 4.48. Birner’s Z-Score is 7.25.

A Z-Score over 3 is considered safe. A Z-Score between 1.81 and 3 is borderline. And a Z-Score below 1.81 means the company is expected to go bankrupt.

Try to use the Z-Score as much as possible. Keep a list of stocks you’re interested in and their Z-Scores. Then make sure you can explain why you should buy a stock with a lower Z-Score. You should make sure you have good reasons to trade down from a high Z-Score stock to a low Z-Score stock because lower Z-Scores are riskier.

I like a portfolio with a higher average Z-Score. But that doesn’t mean I’m going to sacrifice things like 10 straights years of free cash flow. Usually you don’t have to.

Just avoid buying stocks with a Z-Score under 3. And obviously never buy a stock with a Z-Score less than 1.81. That means the company is expected to go bankrupt.

You don’t want to risk that.

Well that’s all for today’s show. If you have a question you want answered call 1-800-604-1929 and leave me a voicemail. That’s 1-800-604-1929.

Thanks for listening.



About the author:

Geoff Gannon
Geoff Gannon


Rating: 4.5/5 (8 votes)

Comments

mattius10
Mattius10 - 4 years ago
In regards to the Asset Turnover ratio in question 5, you reference the previous podcast that was about F-Score. In the F-Score podcast you recommend using "starting assets" to calculate the Asset Turnover Ratio. Is this also the case when calculating Asset Turnover for the Z-Score? Also, are there any other questions where we should use starting assets (as opposed to the current year's total assets) for Z-Score?
Geoff Gannon
Geoff Gannon - 4 years ago
Mattius,

I only mentioned using starting assets for the F-Score, because Professor Piotroski specifically said that's what he used in his original paper.

Often, it won't make much of a difference which asset number you use. I can't link to Professor Altman's original 1968 paper that discussed the Z-Score study methodology, but you can find it online if you have access to a scholarly journal database like JSTOR.

Balance sheets are usually presented with last year's balance sheet there right alongside this year's. So the starting assets would be last year's ending assets. The ending assets would be this year's ending assets. And the simplest average assets for the year is just those two numbers added together and divided by 2. You can check all 3 methods for a few stocks and see how little they normally differ.

Personally, I would use starting assets. It's one less step. But, obviously, never use starting assets when you do a quarter to quarter comparison, only year over year. Assets fluctuate a lot in some industries during the year due to seasonality.

mattius10
Mattius10 - 4 years ago
Thanks Geoff. Appreciate the feedback.
v4value
V4value - 3 years ago
Geoff,

Given that the Total Assets element is so fundamental to the Z-Score, how would you go about treating intangible assets? I've been analysing a UK stock (Armour Group, AMR) and they have a Z-Score of 2.4 with intangibles included and a Z-Score of 3.8 excluding intangibles. Do you have a view on which way to go, because it's fundamental to the outcome?

As always, you're advice is much appreciated.

Regards

Anthony
gnaray
Gnaray - 2 years ago
Geoff,

Do you adjust workiing capital for excess cash when calculating Z score ?

Thanks,

Gopal

Please leave your comment:


Get WordPress Plugins for easy affiliate links on Stock Tickers and Guru Names | Earn affiliate commissions by embedding GuruFocus Charts
GuruFocus Affiliate Program: Earn up to $400 per referral. ( Learn More)
Free 7-day Trial
FEEDBACK