If you follow a major company, you may have noticed that the stock will sometimes move based on an analyst's upgrade/downgrade on the stock. And within that upgrade/downgrade, analysts will often set a mysterious target price for the stock, with no discussion of how they calculated that price. How do they get to it?
In short, they use financial models. Here's how a stock price is broken down, in a very generalized way:
Readers are probably familiar with multiples, which is fairly simple to calculate and pull up on Yahoo Finance. But what about the earnings? Financial models help analysts forecast this component based on their view of what might happen to the company in the future. This is where the meat of the analysis on Wall Street is conducted.
In this post, I'll walk you through exactly how analysts go through a financial model and forecast line items, using Home Depot as an example. I'll discuss line items from my Home Depot Advanced Model, but you can also follow along with the Basic Model and skip over some of the line items that aren't present. Note that this post will be a somewhat technical deep dive into the forecasting process.
The General Forecasting Framework
Let's first talk about the general forecasting process. Many companies give guidance in order to help analysts set expectations. I start with management's guidance (if it exists), since they see recent trends, have more internal data, and have a deep understanding of their business. Management's guidance and commentary can be found in both the earnings press release and the earnings conference call.
From there, I then begin to question management's guidance based on my own opinions about the company/industry as well as historical trends. My own opinions are shaped by my understanding of the underlying drivers of the business, developed over time by following the company and the industry. Historical trends can be found inside my earnings models. If I find a disconnect between my own opinion and management's expectations, I'll then double check my own assumptions to make sure I am not missing anything. If not, I'll forecast to my opinion.
Home Depot's 2015 Earnings Forecast
Now let's take Home Depot as an example and work our way down the income statement.
Retailers forecast sales by calculating sales from their new stores, and sales from their existing stores. An explanation for why the stores are split up in this way can be found in my post on same store sales.
Forecasting Home Depot's new stores is a straightforward process. Home Depot is a mature retailer with stores that are close to saturation in North America. Management noted that they would open 6 stores in 2015, which would represent a measly 0.3% of total store growth. This number could be higher or lower, but it really doesn't move the needle either way, so it's not worth spending much time on.
Same Store Sales
This is the most important input in the forecast because it drives most of earnings. The number you input here will also impact profitability forecasts. Management's guidance is for 3.3% - 4.5% growth, so let's start with that. Does this guidance make sense based on the underlying drivers of the business, based on what I know is occurring at the company, and based on recent trends?
Let's start first with the drivers of same store sales. Home Depot's comps are driven primarily by economic growth and the strength of the housing market (as I've detailed in the past, both housing prices and housing turnover are key metrics that drive home improvement spending). Within the housing market, we know that the industry has recovered well since the housing slump from 07-11, and both turnover and pricing has improved significantly. More recently, turnover has slowed, but pricing remains robust. Overall, the housing market is likely a slight positive. (Every so often companies will attend investor conferences where they'll discuss underlying industry drivers and company initiatives in more detail. Home Depot did just this last week; you should check out the transcript for more details)
In terms of recent trends, Home Depot has seen strong comp growth for the last several years, driven in part by the strong housing industry mentioned above and in part by management's company initiatives to improve customer service (you can read about the company's latest initiatives in their earnings conference calls). Recent trends show that their strength continued in 2014, with the largest comp coming in their latest quarter at 7.9%. Home Depot is also gaining share from Lowe's, as they've had higher same store sales for multiple years.
Thus, looking at managements guidance, 3.3% - 4.5% comp growth looks more than achievable given that 1) it lines up with a slightly positive housing and economic backdrop, and 2) guidance would actually represent a deceleration from more recent trends (and thus appears to be conservative). As a result, I've modeled a comp at the high end of the range for the full year.
As part of the comp growth forecasts, I also have to forecast comp growth by quarter (in the advanced model). Doing so involves consideration of management's commentary, comparisons, and other seasonal factors. Management noted that 1Q comps would likely be the highest for the year, while 2Q comps would be the lowest. There's several factors that align with management's commentary. Comps accelerated through 4Q and were highest in January, suggesting that 1Q could also show strong comps (monthly comp data is located in the Other Metrics section in the basic model and in the Historical Comps tab in the advanced model). Additionally, 1Q has the easiest comparison from the prior year. Meanwhile, 2Q has a more difficult comparison, as 2Q last year benefited from a late start to spring (Home Depot generally sees a significant sales benefit when spring starts, which is usually in 1Q). Thus, I'll model 1Q as the highest comp (at 5.5%) and 2Q as the lowest comp (at 3.0%).
Moving onto gross margin. Management guided to flat gross margin in 2015. How does this stack up to recent trends and what we know? Gross margin represents the companies efforts to manage inventory and obtain goods at cheaper prices. If they buy too much inventory, they'll have to discount heavily. If they buy too little, they'll be out of stock and leave sales on the table. It's a fine balance that Home Depot has gotten better and better at with a strong supply chain that provides items to stores just in time. While this fueled gross margin gains for some time, recent performance shows that they're now starting to hit a wall in gross margin gains. As a result, flat gross margin for 2015 seems realistic, and I've modeled that for the full year.
In general, SG&A expenses are more fixed in nature. Companies plan to spend a certain amount each year, with an assumption for sales growth. Their spending plan includes things like salaries, hiring, IT investments, opening new stores, advertising, etc. Most of this is set early on in the year. If sales start to come in higher than what they've planned, they likely may hire a few more people, but their expenses likely won't change a huge amount. Alternatively, if sales start to fall short of their plans, they won't be able to lower their expenses as much, since much of it was already set in stone at the beginning of the year.
Therefore, a big part of what you forecast for your SG&A (which, remember, is as a percentage of sales) is going to depend on what you've forecasted for sales, since that's the part that's likely to move a lot. If you forecast a high sales growth, then expenses would be a lower percentage of sales. If you forecast lower sales, expenses would be a higher percentage of sales. You should also watch the SG&A dollar growth to make sure that your forecast makes sense. For example, you probably don't want to forecast so much leverage that your SG&A dollar growth is actually declining (unless management said it would decline, or you have a strong reason for thinking so).
Onto Home Depot. Management gives guidance for SG&A expense as a percentage of sales growth. Their SG&A guidance is that it will grow at roughly 40% of total sales growth rates. Historically, Home Depot has been able to do even better than this, as their SG&A expense growth in 2014 was 1/4th of its sales growth rate. As a result of their sales growth and management's tight control of SG&A, expenses as a percentage of sales have leveraged significantly over the last several years, declining by 50-90 basis points each year. With sales growth of 4.5% modeled for the full year, management's guidance would suggest SG&A growth of roughly 1.8%. However, given its history of outperformance, I'll model slightly more SG&A leverage, and SG&A dollar growth of 1.1%.
Other Details (depreciation, share count, tax rate, capex)
The above inputs cover the major forecasts, but I'll detail some of the other line items here. The majority are based on management's guidance because they are largely driven by management's decisions/internal forecasts.
Depreciation: Management gave depreciation guidance of $1.8 billion for the year. Depreciation is a straight forward calculation that management is best positioned to forecast, so my forecast will stick to their guidance.
Share Count: Management gave guidance of $4.5 billion share repurchases for 2015, which is slightly below prior year levels. I've evenly spread out this $4.5 billion each quarter and assumed an average stock price to get to amount of shares that they would repurchase. I then reduced the share count by this amount.
Tax Rate: Management gave guidance of 37%, which is slightly higher than prior historical tax rates, but seems largely reasonable. This might be a slight source of upside if it comes in at historical levels.
Capital Expenditures: This is a forecast on the cash flow statement that will be used to calculate free cash flow. Management's guidance was for capital expenditures of $1.6 billion, which is roughly in line with what the company normally spends in each year as a percentage of their sales.
After going through this forecasting process for 2015, I then repeat the process for 2016. Generally, management teams do not give guidance beyond the upcoming year, so I've modeled similar (but slightly more moderated) trends occurring in 2016 based on my understanding of the company -- 4% comp growth and 50-60 basis points of SG&A leverage.
The Earnings Output
The result of all the inputs is 2015 EPS of $5.17 and 2016 EPS of $5.88. The 2015 EPS is at the high end of management's $5.11 - $5.17 guidance, and would represent 13% EPS growth. After considering all of the above inputs, this seems somewhat conservative and leaves some room for upside potentially coming from comps and SG&A.
Forecasting a Target Price
The next step is figuring out an appropriate multiple to apply to our forecasted EPS. A multiple is meant to be a catch-all for everything that is not captured by your EPS estimate. Here are some of the things that that might include:
- Future earnings growth (beyond the forecasted period)
- Risks to that earnings growth
- Stability and visibility of earnings growth
- Management's ability to execute (which could be a component of any of the factors just mentioned)
- Future competitive threats (a component of risk)
- Current investor appetite for risk
This is a lot to consider and whittle down to one number, so you can start by looking at the company's historical trading ranges. Make sure that you are looking at historical forward PE multiples. You can also look at comparable businesses to get a sense of how investors perceive other competitors. This should give you a good starting point.
Next, take a look at the company's future earnings growth prospects. If you've forecasted out multiple years for the company, you can take a look at the earnings growth within those years. For Home Depot, my earnings forecasts for the coming years are for roughly 13-15% earnings growth in 2015 and 2016. Given that many companies trade at least at a 1.0x PEG ratio, a 15x multiple would be a good place to start. From there, adjust that multiple based on the factors listed above. In Home Depot's case, the company has consistently posted solid earnings growth (with 5 years of 20%+ earnings growth), operates in an oligopoly with a tame pricing environment (Lowe's and Home Depot don't compete fiercely like Best Buy and Amazon do), has a management team that is widely admired, and is heavily driven by a housing market that might still have significant upside. As a result, a case can be made that the multiple should be trading north of 20x (or what would be a 1.5x PEG ratio).
Based on all of the information above, I'll use a 23x multiple and apply that to my 2016 EPS estimate. What I'm left with is a $135 target price for Home Depot.
A Framework for Debate
So that's the (shortened) process that analysts go through to arrive at their target prices. But there's more value to this process than meets the eye.
With all of this established, we now have a framework in which we can debate Home Depot's worth. If bears are arguing that Home Depot's margins are reaching peak levels, you can see how that might slow earnings growth, lower the multiple, and lower Home Depot's stock price. Alternatively, if bulls are arguing that Home Depot has multiple years of 5% comps ahead of them, you can see how that might increase sales, increase profitability, and increase the multiple and stock price. If you want to attack an opposing argument, you can quantitatively debate it by picking the parts of the income statement that you disagree with.
The process also helps you understand press releases and conference calls. If you read the conference call transcript, you'll note that management and analysts often speak to each of the major inputs or drivers of the model. The questions that analysts ask are often trying to get to what they should model for the company going forward.
Finally, you can model out different scenarios to determine the risk/reward. How much downside would there be if the housing industry slows, comps go negative, and profitability begins to decline? How much upside do you have if Home Depot reasonably outperforms expectations? Is the upside worth the downside, with all things considered?
I imagine this post was pretty dry, technical, and probably confusing, so if you're still with me, feel free to shoot me a message if you have any questions about this whole ordeal.