Ever since banks and savings institutions began selling mutual funds and annuities in the early 1980s, there has been tension between the advocates of transforming the bank branch into a financial services store and the executives responsible for the traditional core business of banking—collecting deposits and making loans. By selling investments a bank is essentially converting some spread income from deposits into fee income. Alarmists in the traditional banking camp argue that by selling investments a bank could disintermediate itself out of existence.

Advocates of the “bank as a financial services company” counter that the bank’s customers need alternative investments, and in fact will buy the investments from some other source if the bank does not offer them. Since the bank’s competitors are selling investments, by eschewing investment sales the bank is ceding its customers’ investment business to another bank or securities firm that is also offering banking services like loans and federally insured deposits. By letting customers establish relationships with competitors that also provide banking services, the bank is not just at risk of losing some deposits, but is placing its whole customer relationship at risk.

Also, many banks have been trying to diversify away from the spread business. By building up their fee business, banks can insulate themselves more from cyclical loan demand. Indeed, many analysts will downgrade a bank’s stock if the bank has less fee income than its peers. But the problem is that the fee income is earned in many cases by cannibalizing the bank’s own deposits. Thus, instead of a bank augmenting its spread income with some fee income, the fee income in fact is cutting into the spread income. Deposit products and investment products compete for some of the same dollars.

The advocates of investment sales argue that even though some investment sales cannibalize deposits, the commissions on mutual fund and annuity sales are much larger than the spread income on deposits. But the deposit guardians point out that this spread is earned year after year.

Despite this internal conflict, about half of the banks participating in the annual bank brokerage surveys conducted by Kehrer-LIMRA actively try to convert maturing certificates of deposit into investment sales. But this activity does seem to be cyclical, rising during periods of slack loan demand and shrinking when banks are more fully loaned up.

These annual surveys (the Kehrer-ENSI Financial Institution Investment Program Benchmarking Survey) also indicate that many banks are not particularly concerned with disintermediation. In 2006 63% of the banks reported that they provide customer lists to the bank’s brokers. And 48% do not even monitor the extent to which investment sales cannibalize deposits.

Ups and downs of disintermediation
The annual Kehrer-ENSI Financial Institution Investment Program Benchmarking Study has measured the extent of disintermediation since 1991. The 2005 study covered 84 banks, credit unions, and third party brokerages that collectively accounted for over half of all bank mutual fund and annuity sales.

The benchmarking data indicate that in 2005 34% of the funds used to purchase an investment in a typical bank were drawn on that bank’s interest-bearing deposits.

Over the years, we have observed a general downward trend in this statistic. In the mid 1980s, when S&Ls were just beginning to sell annuities, it was common for institutions to report that 80% of the dollars used to purchase investments came from their own deposits.

 

By the time we began systematically measuring disintermediation, the average bank reported that 60% of the investment sales were funded by their own deposits. This statistic generally declined year by year, reaching a low of 30% in 1999, as deposits were leaving banks in droves, before moving back up after 9-11 to 44% in 2003, as assets rushed back into the safe harbor of bank deposits. In recent years the disintermediation rate has settled back down to about one-third of investment sales [Exhibit 1].

This downward trend suggests that banks have learned how to control the extent of disintermediation as they have gained experience selling investments. But there is not such a clear correlation of the disintermediation rate with program age. Banks that have been selling investments for 13 to 14 years have a deposit disintermediation rate that is 14 to 15 percentage points less than banks with less than nine years of investment sales experience and three percentage points less than banks with between nine and 12 years in investment sales. But the banks that have been selling investments longer than 14 years have among the highest disintermediation rates of the banks in the survey [Exhibit 2, p. 52].

How much leaving anyway?
One issue in measuring the extent to which investment sales disintermediate deposits is that people are using their deposits to invest in mutual funds, annuities, and stocks and bonds whether their bank is selling investments or not. About half of the banks that report disintermediation rates attempt to measure the extent to which money is leaving anyway. This is generally estimated in one of two ways:

1. As part of post-sale customer satisfaction interviews, a sample of customers is asked whether they would have used the deposits to buy an investment from some other source if the bank was not selling investments.

2. As a deposit-retention tactic, tellers ask customers making a significant withdrawal what they plan to do with the funds. If the customer says they are going to use the money to make an investment, the teller encourages the customer to first talk with the bank’s investments sales person in the branch before investing elsewhere. If that referral results in an investment sale, the sale is recorded as “intercepted money.”

The second method is often tracked by the bank for commission accounting, because some banks pay higher sales commissions on investment sales that do not disintermediate deposits. In those cases, investment sales staff naturally insist on credit for intercepting funds that were going out the door anyway.

In 2005 these banks estimated that, on average, 20% of the domestic deposits used to fund investment purchases were intercepted or leaving anyway to purchase investments.

Banks that measure money leaving anyway focus on what we call true disintermediation—the domestic deposits that were used to fund the investment sales less the deposits that were leaving anyway. The true disintermediation rate for these bank averages 24%.

 

Thus the typical banks had a gross disintermediation rate of 34% in 2000, but a true disintermediation rate of only 24%. This is substantially lower than some alarmists believe.

Netting out referrals
Advocates of investment sales argue that disintermediation is even lower, because investment sales staff also refer loan and deposit business to the bank. Their position is that the dollar amount of loans and deposits successfully referred to the bank should be netted out against the deposits transferred to investments.

To what extent do investment sales staff refer customers to the bank for banking products?

In our survey, only 44% of the participating banks say they track referrals back to the institution.       

 

In 2005, investment sales staff in these banks received 3.9 referrals from banking staff for every referral they made to the bank, down from 4.5 in 2004. While the average broker received 178 referrals a year from the bank, that broker made 33 referrals to the bank, less than three per month.

Only 31% of the banks studied track the dollar amount of deposits acquired through referrals from investment staff, and only 21% monitor the loan balances derived from these referrals.

While the flow of referrals from the investment sales staff is relatively small, the dollar amount of new deposits and closed loans is large relative to investment sales. For the banks that track this benchmark, the dollar amount of new deposit accounts opened is equal to 16% of the banks’ investment sales, on average. For the smaller number of institutions that are able to report the impact of referrals from investment sales staff on closed loans, the new loan balances amounted to 6% of those banks’ investment sales.

We can bring all of these factors together to estimate the overall impact of investment sales on deposits on a pro forma basis. For the banks that track intercepted funds, 30% of investment sales are paid for by drawing down the bank’s own deposits. But 20% of those disintermediated deposits are estimated to be leaving anyway, by the banks’ own reckoning. Thus true disintermediation is only 80% of 30%, or 24% of investment sales. If we subtract the loans and deposits that result from the referrals from investment sales staff, which amount to 22% of investment sales in the banks that track these benchmarks, we are left with a negligible amount of net disintermediation—about 2% of investment sales.

When we performed the same calculations for 2004, we found that net disintermediation was slightly negative, i.e., investment sales that year actually intermediated deposits somewhat.

Impact on a bank’s deposit base
The disintermediation rate is the share of a bank’s investment sales that are funded by the bank’s own deposits. That is not the same as the percentage of a bank’s deposits that are disintermediated by investment sales, because the annual investment sales in a bank tend to be a small fraction of a bank’s deposits. For example, in 2005, bank investment sales averaged 3.2% of bank retail deposits, but gross disintermediated deposits averaged less than 1% of a bank’s retail deposits (30% of 3.2%). The true disintermediated dollars, after accounting for money that was leaving the bank anyway, amounted to just 0.77% of the average bank’s retail deposits (80% of 0.96%). Netting out the deposits brought in by referrals from investment sales staff reduces the disintermediated deposits to just 0.26 % (0.77% minus 16% of 3.2%). And netting out some value for the loan balances acquired reduces the drain on a bank’s deposit base to negligible levels [Exhibit 3].

Other issues examined by the full study include:
To what extent do customers who buy investments where they bank replenish the disintermediated deposits over time?

What is the relative profitability of investment sales and bank deposits?

What factors influence the extent to which investment sales cannibalize deposits?

What are the ways that can help a bank better control disintermediation? BJ

   

The electronic version of this article available at: http://lb.ec2.nxtbook.com/nxtbooks/sb/ababj0607/index.php?startid=50

TrackBack URI for this entry