The Before and After of Competition
After we determine which banking markets are involved in a proposal, we have to examine how each market will change if the proposal is allowed to proceed. To make this determination, we need to know what each market looks like before and after the combination. We start by building a picture of each market using bank deposit data.
We mentioned earlier that checking or other transaction accounts are the primary accounts customers have with their banks. From this information, we can infer that deposit information is a reasonable measure of a bank’s presence in a market. Using branch-level deposit data, we can calculate a bank’s total deposits and share of deposits in a market. For thrifts, we normally include only half of their deposits because thrifts do not offer all of the same products and services that banks do, particularly to businesses. In other words, thrifts are not “perfect substitutes” for banks. If a bank holding company owns several banks in the same market, then the deposits of the sister institutions are pooled together to determine the bank holding company’s market share. Finally, credit union deposits are not normally included in a market’s deposit calculation. Being membership organizations, credit unions offer their products and services only to certain groups of people, and these products and services are often quite limited when compared with those offered by banks and thrifts. That said, we may include a particular credit union’s deposits in the calculation if substantial evidence supports their inclusion. One piece of such evidence would be that the credit union offers a wide range of consumer banking products. In addition, the credit union should have liberal membership rules (typically, at least 70 percent of market residents must be eligible for membership), and it should have easily accessible street-level branches.
Once we have market shares for all institutions in the market, we can take the next step and determine the market’s concentration. To do this, we use a tool called the Herfindahl-Hirschman Index (more commonly referred to as HHI).
Merger Guidelines
The Justice Department divides the spectrum of market concentration as measured by the HHI into three regions that can be broadly characterized as unconcentrated (HHI below 1,000), moderately concentrated (HHI between 1,000 and 1,800), and highly concentrated (HHI above 1,800). For a banking transaction not to require stricter economic scrutiny in a particular market, the transaction cannot both increase HHI by more than 200 points AND result in a highly concentrated market (a final HHI greater than 1,800 points).
|
To calculate HHI, we simply square all the market shares (expressed
as percentages) and add up the squared numbers. This sum is a
number between zero and 10,000: The smaller the HHI number is,
the less concentrated the market is (the more competition there
is among banks in the market), and the less likely any one bank
is able to exert much control in the market. For example, if a
market has only one bank, it would have a 100 percent market share,
and HHI would equal 10,000, or 1002 x 1. If, instead,
there are 100 banks in a market with 1 percent market share each,
HHI would then equal 100, or 12 x 100. To make this
calculation even easier, CASSIDI® performs it for you for
any banking market in the nation. (See sidebar
on CASSIDI®.)
To determine if a deal will satisfy the antitrust requirements, we need to look at the buyer’s market share after the transaction and the market’s HHI before and after the combination. If the “after”-market HHI is not above a certain level and the increase in the market HHI caused by the deal is not above a certain level, then the deal satisfies the Justice Department’s merger guidelines. (See text at right.) Specifically, the department generally will not challenge a banking proposal unless the after-market HHI exceeds 1,800 points AND the increase in HHI resulting from the deal exceeds 200 points. This is often called the “1,800/200” rule and is unique to banking. Other industries are allowed only a 50-point increase in HHI when it is above 1,800 points. The difference is that the Justice Department recognizes that banks face competition from a variety of other financial providers, such as thrifts, credit unions and other types of financial firms. Allowing banking markets the leeway of a 200-point change in HHI accounts for the “expanded” competition banks face.
In addition to the Justice Department’s rules, the Federal Reserve also will typically not allow a bank to buy its way to more than 35 percent of any banking market’s total deposits. Although similar in structure to the national- and state-level deposit share caps mentioned earlier, this market-level threshold is a Federal Reserve policy, not a law, and exceeding it triggers a closer examination of the market’s economic circumstances, not rejection of the proposal. As with the national- and state-level caps, banks can grow their way to controlling more than 35 percent of a market’s total deposits.
What if the Picture Is Not Clear?
If one or more of the banking markets in a transaction do not satisfy the 1,800/200 rule, does that then mean the transaction cannot go through? No, not automatically. What it does mean, though, is that we will need to investigate those markets further to find out if perhaps other important factors aren’t being picked up by the HHI calculation. One of the first items we’ll look at is the number of other banks remaining in the market and each one’s market share after the deal. We’ll also want to know if any new banks have opened in the market recently and if deposit, income and population growth in the area have been relatively strong when compared with similar areas in the rest of the state. We’ll look to see if a thrift in this market has been aggressively pursuing business customers, making its share of loans to businesses look more similar to other banks than to other thrifts. If so, we may end up including all of that savings institution’s deposits rather than just half in the market’s deposit calculation. Or, there may be a credit union in the market that has a storefront like a bank or thrift and opens its doors to most people in the area. If so, we may end up including a portion of its deposits in the market’s deposit calculation. We would also need to know if the bank being bought is in trouble, perhaps even on the verge of shutting down. We can then use some or all of this information to demonstrate that factors are at play in the market that are not being captured by the HHI, and, when these factors are considered, the deal will not end up substantially lessening competition in the banking market.
At times, though, a market’s current concentration and the potential increase from a deal are just too large for some of these other economic factors to overcome. In such cases, the buyer may offer (or we may require the buyer) to sell branches to other banks in an attempt to keep a local market’s HHI increase to below 200 points. This process, known as “divestiture,” has become increasingly common over the past decade or so, and many institutions, particularly those engaging in large transactions, now come to the table with divestiture plans already laid out.
Competitive Analysis In Action—The Real World
To get a feel for how a competitive analysis might actually play out, let’s look at a recent real-world acquisition—Regions Bank’s purchase of AmSouth Bank. Before the deal, Regions was the 21st largest bank in the nation (based on total assets) and controlled less than 1 percent of national deposits. It operated branches in 16 states. AmSouth was the 27th largest bank in the country and also controlled less than 1 percent of national deposits. It operated branches in seven states. After the deal, Regions became the 13th largest bank in the country and controlled less than 2 percent of national deposits.
Regions and AmSouth had branches in 67 common banking markets across seven states: Alabama, Florida, Georgia, Kentucky, Louisiana, Mississippi and Tennessee. A competitive analysis like that described above was conducted for each of these 67 local banking markets. In 42 of them, the 1,800/200 rule was satisfied without divestitures or any further market examinations. That left 25 markets in which the 1,800/200 rule was violated and/or Regions’ after-market share exceeded 35 percent. Each would require divestiture, further examination or both. In 12 of these 25 banking markets, divestitures of AmSouth branches were enough to satisfy the 1,800/200 rule.
The remaining 13 markets required further examinations because, even after accounting for any proposed divestitures, they fell outside the 1,800/200 guidelines and/or Regions’ after-market share exceeded 35 percent. Credit-union deposits played a role in countering the initial HHI analysis in 11 of these markets, and thrifts in three markets were considered full competitors with commercial banks. In addition, new bank openings in the recent past, strong income, population and deposit growth relative to surrounding areas, and the number and strength of the remaining competitors in all 13 markets contributed to the final decision to approve the application. Thus, the initial HHI analysis did not fully explain the actual competitive picture in these markets. When all was said and done, the information gathered and actions taken were sufficient to conclude that the deal would not have a significantly adverse effect on competition in any of the banking markets. The acquisition was approved in October 2006. Read more about the outcome of this case at www.federalreserve.gov/boarddocs/press/orders/2006/20061020/attachment.pdf.
back to top | next>>
