BRANCHING IN THE "NEW PARADIGM"

Yesterday's article on Sprinkles not only made more than a few bankers hungry, it also brought up the question of where should you put branches that handle both business and retail traffic in this environment? The question is a good one, so we went to work fine tuning our model to put a greater weight towards loan growth - since that has become more of a determinant to profitability relative to deposit and fee growth. We tweaked our model so that loans (including credit risk) have a much greater weighting and point out the fact that a commercial or hybrid (a combo) branch model is very different than the traditional retail branch model. We then looked at socio and demographic trends (growth, age, etc), branch formation/closure, banking statistics (deposit/loan rates, sensitivity, etc.) and economic data (housing, unemployment, loan delinquencies, income, etc.) by MSA from 2011 out to 2016 (to determine where the optimum place to expand is). The first takeaway from the analysis is that branching is about 50% as profitable as it once was, due to margin compression on both the loan and deposit side, combined with a cost structure that really hasn't changed (in fact it appears to have risen on a per transaction basis). Only about 60% of branches are profitable and the ratio of CDs to non-interest deposits is one of the largest determinants of profitability after size. In addition, a lower amount of fee income further hurts branch profitability. Another less obvious takeaway is that the unemployment rate has even more of an impact on branch profitability (because of credit) and now accounts for almost 55% of performance vs. about 45% when we looked at this issue back in 2006. For the most part, a lack of profitability is compounded by the fact that while total branches declined by 600 stores (1.5%), and despite some branches getting a huge increase in deposits; deposits barely increased for the average branch in 2011. In fact, about 30% of branches actually lost balances (actual calculation is difficult as it depends how you allocate balances at main branches). Still, despite these factors, 90% of banks didn't change their branch structure last year. Those that did were usually large banks (Wells Fargo, BofA and Regions were the bulk of the market). Now that you have the background out of the way, let's look forward and see some quality markets for future branching. First, because of demographic trends, metro branches have surpassed rural branches as being more profitable. Of the 140+ markets that we looked at, the top 5 winners were Los Angeles, New York, San Francisco, Washington D.C. and Miami. A couple of notables were that almost any metro area in CA was a good one (Sacramento, San Jose, San Diego, etc.), as competition is the lowest in the country and credit has (and is expected to) improved the fastest. Other strong areas that surprised us included Honolulu (low sensitivity), Boston (affluence), Raleigh (many factors) and Minneapolis (many factors). Finally, in a new twist to our analysis, we included start up/operating costs (labor, rent, insurance, etc.) and future expected credit formation and asked the question - what is the best place to have a branch with the lowest risk (upside to downside)? Surprisingly, a cost structure and the upside potential in lending make Las Vegas, Provo, Austin and Phoenix some of the lowest points of entry with the best upside. Branching is still more art than science and while we are experts at bank profitability and economics, there are many firms far more qualified to discuss socio and demographic trends. So, don't take our analysis as anything more than trying to get you in the right ballpark, while bringing up some issues to think about. The products you offer and your objectives can radically change branch decisions, so our analysis is more generic in construction. However, since geography has such a major impact on profitability, there are clear winners out there that still give your bank the best chance of having a profitable new branch.