Shares in Wells Fargo (WFC) were down as much as 2.5% on Friday, before closing off 1%, after the release of quarterly earnings in which some investors were disappointed because of the net interest margin (i.e., the difference between the interest rate received on interest-earning assets, such as loans, and the interest rate paid on interest-bearing liabilities, such as customer deposits).
Here, for example (paid subscription), is what Mr. Jim Cramer had to say:
Net interest margins fell 10 basis points to 3.56%, which was in line with analysts' expectations and better than the 24 basis points drop last quarter. But we wrote yesterday we needed 3.67% or better.
He went on to clarify that, as a result, the quarterly report was
not good enough to lead to higher earnings revisions and to support the strength in the shares.
In this article, we explain how the institutional dynamics of Wall Street give rise to such an extraordinarily precise assessment of WFC's quarterly report, and how retail investors with a sensible long-run time horizon can take advantage. Specifically, we see the weakness in WFC as an opportunity. This is because the value of the shares, to someone who views them as a long-term ownership interest in the underlying business, is utterly unaffected by where the net interest margin in the just-reported quarter falls in a ~10 basis point range (i.e. between the reported 3.56% and the 3.67%, which Mr. Cramer believed was "needed for a higher stock price").
The Earnings Models of Analysts
Research analysts on the "sell-side" (i.e. working for brokers such as Goldman Sachs (GS) or Morgan Stanley (MS) who sell stocks to institutional "buy-side" clients, such as mutual funds and hedge funds, who manage money) expend a great deal of time attempting to model on a spreadsheet the earnings of the companies they follow.
An important output of these earnings models, which will include revenue and expenses on a quarter-by-quarter basis at least one or two years into the future, is the earnings estimate for the current year and the next year. For example, taking the average or "consensus" of analyst estimates reported by Yahoo Finance, WFC is expected to generate earnings-per-share or "EPS" of $3.61 in 2013. In the next month or so, those professional analysts who have not already done so will roll forward their forecasts, giving rise to a consensus estimate for 2014.
Beware EPS Estimates
The pitfalls of this EPS estimation exercise, particularly for bank earnings, are evident if you think about attempting to forecast your personal or family budget over the next two years. Even supposing you get the basics right, such as your wages and home expenses (e.g., rent or mortgage and utilities), how are you going to forecast the return on your savings account or the interest cost of your credit card? What if you lose your job, or get hurt and face medical bills, or get divorced? What assumptions will you make for tax and health insurance costs?
This is not to say the exercise is futile. Rather it is the process, and the impact it has on the present management of your finances, that is important, and not the output estimate of how much excess income you may or may not generate in 2014. And yet, on Wall Street, it is the consensus of these estimates that can drive stock prices. In the particular case of WFC today, as often happens, things can become baroque: It was investor expectations of analyst expectations of WFC's future earnings that appear to have accounted for the stock price action.
Disappointed with the forecasting accuracy of your family-budget model, you decide to dig deeper into the numbers. Rather than model your income in aggregate, you break it into component parts: the salary from your day job, the moonlight income from writing for Seeking Alpha, and the realized gains from your stock portfolio. (Unrealized gains are important but you cannot buy groceries with them).
If you were a professional sell-side analyst covering WFC, the equivalent would be unpacking your revenue forecast into "line items" such as net interest income, fees, and realized gains such as those on "available-for-sale" securities that management has indicated it does not intend to hold to maturity. (Unrealized gains are important but flow to the balance sheet through other comprehensive income, not earnings).
There is something reassuring about seeing these carefully-labeled numbers marching neatly across the quarters on your spreadsheet. So, you go deeper and unpack the income from Seeking Alpha into an estimate for the number of articles you write, an estimate for the page-views per article, and an estimate for the price paid per page view. You are not sure how many people will read your articles but you anchor your estimates for the future to your current popularity with some assumptions for growth and/or variation over time.
If you were a professional sell-side analyst covering WFC, the equivalent would be unpacking your forecast for net interest income into an estimate for the dollar-value of loans and other interest-earning assets, and an estimate for the net interest margin. You are not sure how the net interest margin will evolve but you anchor your estimates for the future to the result in the last-reported quarter with some assumptions for growth and/or variation over time.
This approach has questionable consequences. With estimates for the future anchored to the last report, a decline in the net interest margin in the current quarter from the last quarter downshifts results across the entire forecast period of the model and so affects the earnings forecast far into the future.
Of course, the process is not entirely mechanical and diligent analysts will take account of one-time effects and mitigating circumstances. But, in the case of WFC, the dynamics of the spreadsheet model created a bias if the margin for the quarter was below the prior quarter result. Hence, Mr. Cramer's view that if WFC's net interest margin was not 3.67% or above (versus 3.66% in the last quarter) analysts would find it difficult to raise their estimates.
Exaggerating for effect, it is as if an author on Seeking Alpha were to constantly revise a forecast for page-views per article in late 2014 based on the result from the most recently-published piece. And then make a decision about how much time to devote to writing in the future, based on that revision!
I do not know what the net interest margin for WFC will be in 2014. It will depend on interest rates (and, more specifically, the yield curve) and a reasonable analysis might start by making a projection for the yield curve based on the "forward" rates which represent the expectations of the bond market.
I do know, however, that these future interest rates, and even the bond market's expectations for them as reflected in the forward rates, are not impacted in the least by the fact that, in this just-reported quarter, WFC saw its net interest margin fall ~10 basis points from the last quarter.
However, this "sequential" margin decline influenced the perception of some investors for the dynamics of analysts' earnings models and hence their assessment of the chance of an increase in earnings estimates. The stock appears to have traded negatively as a result, creating an opportunity.
Disclosure: I am long WFC.