New York | In the steamy days of July 2018, bank earnings are more interesting than usual, in part because our beloved friends and colleagues in the financial world continue to intone the false mantra that rising interest rates are good for banks. And our former colleague at Bear, Stearns & Co many years ago, David Solomon, took the baton at Goldman Sachs (GS). Break a leg David.

First off, it is not true that rising rate rates are always good for banks. Increased interest rates, for example, have not helped GS in the past few quarters. "Banks make money on the spread, that's it. That's the story," said our friend Josh Brown on CNBC’s Fast Money this week. At the time, Brown was surrounded by a bunch of earnings happy pundits singing the praises of higher short-term interest rates for bank earnings.

Banks make money on widening spreads, not because of a quarter point move in Fed funds. Spreads, of course, are not really widening as the Federal Open Market Committee pushes up short-term rates. After moving up about 75bp points over the past year, non-investment grade spreads started to fall after the most recent FOMC rate hike in June 2018. Indeed, the 10-year Treasury has declined about 30bp in yield over the past month, reflecting the continued tightness of credit spreads – at least for some issuers.

Second comes the earnings themselves. The Q2’18 bank results released so far confirms the accelerating upward trend in bank funding costs. As we detail in the most recent, Q2’18 edition of The IRA Bank Book, the rate of increase in funding expense for all US banks should exceed the rate of growth in interest earnings by Q2’19. This will mean that net interest margin or “NIM” will be shrinking, an eventuality that most market analysts and institutional wealth managers really are not prepared to accept -- much less reflect in asset allocations.

The extract below from The IRA Bank Book shows our projection for aggregate funding cost through Q4’19. Notice that we hold the growth rate in total interest income steady at 8% through the end of 2019, an admittedly generous assumption given the way that the FOMC has capped asset returns via QE. On the other hand, we limit the annualized rate of increase in funding costs to “only” 65%, a rate of change already more than reflected in the rising funding costs of names like First Republic Bank (FRC) and Bank of the Ozarks.

Source: FDIC, WGA LLC

Most of the largest US banks that reported earnings this week saw interest expense rise by mid-double digits even as interest earnings rose by single digits. Goldman Sachs, for example, saw its funding expenses increase 61% year-over-year (YOY) in Q2’18 while interest income rose just 50%. Citigroup (C), on the other hand, being already positioned in the world of institutional funding, saw interest expense rise only 28%. But the Q2’18 earnings seem to confirm a rising trend in funding costs that could see NIM flatten out and decline by 2019.

When Solomon’s ascension to the top spot was announced at Goldman Sachs, our friend Bill Cohan commented on CNBC that this amounted to a takeover of GS by alumni of Bear, Stearns & Co. God does have a sense of humor. He also reminded Andrew Sorkin et al on Squawk Box that the freewheeling Goldman of old is long gone and that GS is now run and regulated as “a bank.” Well, no, not really.

Goldman Sachs is basically a broker-dealer with a small bank in tow. When you compare the net interest margin of GS with its peers, for example, the other members of Peer Group 1 defined by the FFIEC reported NIM of 3.28% vs 0.41% for GS in Q1’18. Because the bank unit of GS is so small, the overall NIM for the group is 1/10th of its peers compared with total assets. Goldman makes less than 2% on earning assets vs almost 4% for its asset peers. So to paraphrase the wisdom of Josh Brown, GS does not make money on interest rates, up or down, but rather earns fees from trading and investment banking. GS profits from the spread, both in terms of price and volume.

The basic problem confronting David Solomon and his colleagues is that GS really is not a bank. It is regulated like a bank and therefore constrained in terms of business activities, but it does not earn the carry on assets that most banks take for granted when they turn on the lights each morning. Talk of expanding the banking side of the business (aka “Marcus”) is fine, but progress in this regard is very slow indeed. Of the $9.4 billion in net revenues reported in Q2’18, just $1 billion represented net interest earnings.

The gross yield on GS’s loan book (5.24%) is superior to its larger peers (4.68%), but the numbers are so small that they are not really significant in the overall picture. The total return on earning assets for GS at 1.85% is less than half of the 3.94% earned by its larger peers. Why the poor performance? Because GS pays up for non-deposit funding compared to its larger peers. Because it has such a small deposit base, Goldman’s total cost of funds is more than twice (2.45%) that of its larger bank peers (0.97%) as of the end of Q1’18.

Organically growing GS into a true commercial bank will take time and a lot of work, a task that Solomon et al may or may not be able to accomplish. Building a bank starts first and foremost with stable funding in the form of customer deposits, particularly commercial deposits from small and mid-size businesses. With the intensifying competition for bank funding now very visible in the money markets as the FOMC shrinks reserves, don’t hold your breath waiting for GS to transform itself into a traditional depository. Such a transfiguration is possible, but not very likely in today’s markets. Thus the question for David Solomon and his colleagues: Do you really want to be a bank? Really?