When and Why Do Banks Compete for Deposits?
James Keenan
I believe that one unsolved problem facing wider acceptance or application of Modern Money Theory (MMT) is our inability to provide a clear, succinct answer to the question, When and why do banks compete for deposits?
The Question Posed by MMT
Note: What follows is likely an oversimplification for rhetorical purpose.
MMT asks and answers a different question or, actually, two different questions:
Do banks lend from reserves?
No, MMT responds, they don't lend from reserves. Bank reserves, held at the central bank, are used to settle balances with other banks.
Do banks lend from deposits -- and, by implication, do banks need to amass deposits before they can lend?
No, MMT responds, it's the other way around: Loans create deposits.
When a bank extends a loan to a customer, the amount of the loan is entered on the bank's books once on the asset side (the loan) and once on the liability side (the demand deposit whose balance is incremented by the amount of the loan).
The customer likewise enters the loan amount once on the asset side of her books (her checking account) and once on the liability side (what she owes).
The total amount of the bank's deposits at the moment the loan is made has no inherent effect on whether the bank proceeds with the loan or not. They don't "lend from deposits"; hence, they do not, as a strict matter of logic, need to amass deposits before they can lend.
The Question Posed in Response
Well then, you might respond, If banks don't need to amass deposits before they make loans, why (at least some of the time) do they compete fiercely for deposits?
When I was young, many decades ago, my uncle would come to dinner on Sunday night and bring along a copy of the Sunday New York Times. The Business section of the Times would have many display ads from savings and loan associations, mainly in California, that would solicit time deposits that paid higher interest rates than you could get at S&Ls in New York state. California was still rising, people moving there needed houses, and the subdivisions were sprawling across the San Fernando, San Gabriel and Antelope valleys. The California banks appeared (to my pre-MMT eyes) to need deposit inflows from people anywhere in the U.S. in order to fund that housing boom. To attract those inflows, they offered higher rates than you could get at, say, First Federal Savings and Loan Association of Rochester.
But it wasn't just S&Ls in California, of course. Commercial banks and S&Ls would periodically go into "toaster wars" in which they would offer people "free gifts" for moving their accounts to their institutions.
So there was competition for deposits? But why, if the scope of bank lending was not limited to the size of pre-existing deposits?
Enter Richard Murphy
British political economist and public intellectual Richard Murphy encountered this topic on his blog on August 17. He was taking note of press reports that, notwithstanding the Bank of England's August 5 interest rate hikes, very few British banks or building societies (the Brit equivalent of American S&Ls) had yet to start paying out higher interest rates to their depositors. Murphy stated that the "real reason" for this is:
... the explanation provided by modern monetary theory. Since banks do not need to hold deposits from customers to make loans, because all loans are made as a result of new money creation, of course banks are not willing to pay people to deposit funds with them anyone. Why on earth would you want to pay for something you no longer need?
How I Responded to Richard Murphy
I wrote a response citing my childhood memories of bank competition for deposits. I added:
Granted, I don’t see much of this competition for deposits these days — but how do we square past competition on interest rates for deposits with what MMT has to say? (Note: many of the people in my MMT classes are old enough to remember this competition for deposits.)
Murphy Responds to Keenan
I've been following Murphy's blog, Tax Research UK, for years, have taught from his book The Joy of Tax and gave that book a plug when he spoke via Zoom at the Levy Institute Summer Seminar this year. Murphy has thousands of people reading his blog and his Twitter Feed, but he can get quite ... tetchy with people who disagree with him. I was on the receiving end of his irritability when he responded:
"The world has changed
"We now know how fiat currencies work
"We now get MMT
"So why suggest outdated thinking says I am wrong?"
I don't believe that I was suggesting that anything Murphy said as a unique individual was wrong. I was suggesting that my students and I could not articulate an MMT-grounded explanation of why banks compete for deposits or, more precisely, why in certain historial periods they compete heavily for deposits and in other periods they do not.
I suspect that this is not so much a problem in the fundamentals of MMT as in its pedagogic transmission to the public. Nonetheless, I believe it qualifies as an unsolved problem in Modern Money Theory.
mgraves mstvp.com
James,
The answer is quite simple. Banks no longer compete for deposits, and haven’t for quite some time.
The possible exception being startups, like when ING entered the US market back in the 90s. That’s about marketing, not economics. They are able to differentiate themselves through paying higher rates on simple deposit accounts. As a practical matter, the sums involved are trivial.
This is similar to the question of why sell treasuries? They’re not an economic requirement. They’re a legal requirement, based upon a law established in the era of the gold standard.
Michael Graves
--
You received this message because you are subscribed to the Google Groups "Modern Monetary Theory" group.
To unsubscribe from this group and stop receiving emails from it, send an email to
modern-monetary-t...@googlegroups.com.
To view this discussion on the web visit
https://groups.google.com/d/msgid/modern-monetary-theory/d84654e0-91e8-460b-b4d2-64997b32e13fn%40googlegroups.com.
James E Keenan
> James,
>
> The answer is quite simple. Banks no longer compete for deposits, and
> haven’t for quite some time.
>
*no longer* compete for deposits [emphasis added] -- suggests that they
once *did* compete for deposits. That begs the questions: Why, at that
past time, *did* they compete for deposits? And what changed such that
they no longer do so?
Those questions are not strictly-MMT questions. If there's been
research into them anywhere in the macroeconomics or money and banking
literature, I'd be happy to take a look. But I'd be particularly happy
if you could point to any place in the MMT literature that deals with
this topic and that I might have overlooked.
ryan.be...@gmail.com
Hi Jim,
Banks compete for customers’ deposits because they are generally cheaper than borrowing in the federal funds market. The current target range for FFR is 2.25%-2.5%, while according to bankrate.com average rates for retail savings accounts are around 2%.
When a depositor instructs funds to be moved from bank A to bank B, the Fed debits bank A’s reserves and credits bank B’s reserves. As of March 2020, banks no longer have minimum reserve requirements, but they aren’t allowed to overdraft/run a negative reserve balance, so they are incentivized to attract customer deposits. Additionally, legislation added in the Dodd-Frank Act established minimum “retail” deposit requirements.
I found a blog post from 2014 that explains more here (with links to pieces written by Warren Mosler and Scott Fulwiler).
I’d like to push back on the contention that banks no longer compete for deposits. In fact, Mosler wrote back in 2010 that Dodd-Frank should drop the minimum requirement for “retail” deposits, because doing so raised the cost of funds for small banks, as it forced them to more aggressively compete for such deposits.
From a business perspective, it may have made sense in the past to offer customer incentives to attract more retail deposits, such as toaster ovens. However, banks not only want to attract customers’ deposits, they want to keep them for as long as possible, as doing so provides a stable funding base, and they have the opportunity to cross-sell higher margin services to those customers in the future. In recent years, banks have found other ways to attract and retain retail deposits, including offering services such as mobile apps with check-clearing technology. Spending on such services is a cost, but if it helps attract and retain more retail deposits, it pays for itself.
Hope that helps.
Ryan
James E Keenan
> Hi Jim,
>
> Banks compete for customers’ deposits because they are generally cheaper
> than borrowing in the federal funds market. The current target range for
> FFR is 2.25%-2.5%
> according to bankrate.com
> <https://www.bankrate.com/banking/savings/rates/> average rates for
>
rooted in actual dollar incentives to banks (and for banks those are the
only incentives that matter ;-) ); and (b) is empirically testable and
historically researchable.
> When a depositor instructs funds to be moved from bank A to bank B, the
> Fed debits bank A’s reserves and credits bank B’s reserves. As of March
I guess that escaped my attention in the early days of the pandemic.
It's discussed here:
https://www.federalreserve.gov/monetarypolicy/reservereq.htm
> but they aren’t
> allowed to overdraft/run a negative reserve balance, so they are
> incentivized to attract customer deposits. Additionally, legislation
> added in the Dodd-Frank Act established minimum “retail” deposit
> requirements.
>
> I found a blog post from 2014 that explains more here
>
> I’d like to push back on the contention that banks no longer compete for
> deposits. In fact, Mosler wrote back in 2010
> because doing so raised the cost of funds for small banks, as it forced
> them to more aggressively compete for such deposits.
>
that it, too, is written in plain English, it's also good because in it
Mosler makes arguments that are empirically testable. Example:
"The primary reason for the high cost of funds is the requirement for
funding to be a percentage of the 'retail deposits'. This causes all the
banks to compete for these types of deposits."
While I would generally trust Mosler on this question, what's really key
about this statement is that other people knowledgeable about bank
operations could submit it to scholarly scrutiny.
> [snip]
Thanks for your response.