By Robert P. Murphy
When explaining the relative safety and stability of the insurance sector, proponents of Nelson Nash’s “Infinite Banking Concept” (IBC) will often point to the 1930s. They make claims that although thousands of banks failed, no insurance policyholders missed a payment.
Is this true? In the present article I’ll rely on a hostile article, with at least one of the authors affiliated with Citicorp, to see just what happened.1 As we’ll see, even though the authors of the piece, Huertas and Silverman (H&S), try to paint a different picture, their own statistics and storyline show that the insurance sector was much more reliable during the Great Depression than the commercial banking sector.
THE ROARING TWENTIES
H&S provide some interesting statistics to show the strong growth of both banking and insurance during the 1920s, which highlight the prominence that the insurance sector used to enjoy. According to H&S:
“The assets of all commercial banks rose from $43.7 billion in June 1921 to $62.4 billion in June 1929, an annual compound rate of growth of 4.5 percent. The assets of life insurance companies grew more than twice as rapidly, from $7.9 billion in December 1921 to $17.5 billion in December 1929, an annual compound rate of growth of 10.4 percent. Total life insurance in force jumped from $43.9 billion at the end of 1921 to $102.1 billion at the end of 1929, an annual rate of increase of 11.1 percent.”
Thus, by 1929, total assets held by life insurers were about 28 percent as much as the total assets held by the commercial banks.
H&S go on to report that the number of actual life insurance policies in force rose from 70 million in 1921 to 123 million in 1929, which was roughly the size of the total U.S. population at the time.
IT’S NOT A WONDERFUL LIFE: RUNS ON THE BANKS
By their very nature, fractional-reserve banks are vulnerable to “runs,” in which depositors seek to withdraw their funds en masse. This is because the commercial banks take in, say, $1000 in cash as a deposit, but only keep, say, $100 in the vault as a reserve. The other $900 can be invested or lent out to another bank customer.
Fractional reserve banking allows the banks to pay interest on demand-deposit (i.e. checking) accounts, but the accounts are thus vulnerable to a run. If the original depositor—who thinks he has $1000 in his account—wants to take out his money, the bank should be able to accommodate him under normal circumstances. Even though the bank has lent out $900 of his initial deposit, there are plenty of other customers’ deposits sitting in the vault, and so the bank can dip into those funds to pay the man his full $1000.
Of course, the problem with a bank run occurs when many customers show up at the same time. In our example, if the bank has only been keeping 10 percent of each deposit in the vault as a reserve, then if customers collectively want to withdraw more than 10 percent of the total amount on deposit, the bank will fail.
As Carlos and I explain in our book, How Privatized Banking Really Works, many Austrian economists are opposed to fractional reserving banking per se. They view it as fraudulent and economically disruptive. Because of the evolution of financial practices and legal rulings, it is now the case that commercial banks can literally create money out of thin air when granting new loans.
The Austrians who follow in the tradition of Murray Rothbard stress that banking doesn’t have to be like this. Commercial banks could distinguish between the functions of (a) warehouse and (b) credit intermediary, by offering different types of accounts. A true 100 percent reserve checking account would require a small fee from the customer, but would otherwise be a perfectly secure way to enjoy the conveniences of safekeeping large amounts of money, and being able to spend it via check or debit card. On the other hand, if the customer wanted to earn interest, he’d have to lend the bank money by putting it into a genuine savings account (or by buying CDs), where he couldn’t access the money immediately.
In the early years of the Great Depression, there were three great waves of commercial bank failures. I should stress that (as usual) government intervention played a large role in this outcome, beyond the existence of fractional reserve banking. In particular, “unit banking laws” greatly restricted the ability of banks to engage in branch-banking in different U.S. states. Therefore, if one region (e.g. a community highly dependent on loans to farmers) suffered major investment losses, the banks in that region would go down, because they were not tied to a larger, national institution. To get a sense of the importance of this fact, note that not a single bank failed in Canada during the Great Depression, arguably because branch banking wasn’t restricted in our neighbor to the north. (For more details, see my book on the Great Depression.2)
When commercial banks began failing, customers of other banks became nervous and began withdrawing their funds too. However, even here I want to mention that it wasn’t simply a free-for-all; research suggests that the banks that failed typically really were in financial trouble. In other words, it wasn’t simply an irrational public rushing to get their cash, but rather that people would catch wind of the fact that their particular bank was in trouble and then they’d run— thus sealing the bank’s doom.
One of the very first acts of the newly installed Roosevelt Administration was to intervene in this process. (Note that at that time, presidents were inaugurated on March 4.) Here’s how H&S describe it:
“This was the Banking Holiday of 1933. As one of its first acts, the new Roosevelt Administration on 6 March 1933 closed every bank in the country. Congress then hastily passed the Emergency Banking Act on 9 March, validating the President’s action, extending the holiday, and empowering the President to license banks to reopen when they were found to be in satisfactory condition. Such banks were allowed to reopen on 13 March in the reserve cities and on 15 March in other places. However, 2,100 banks never reopened at all, bringing the total number of banks that failed during the Depression to 9,100, or 38 percent of the number of banks in existence in June 1930 before the collapse began.”
Clearly this was an abysmal performance for the U.S. commercial banking sector, though to repeat it is unclear what would have happened had there been no restrictions on nationwide banks opening local branches in various states. And naturally, if the banking sector had operated on Rothbardian principles—i.e. where checking accounts were backed up 100 percent—then there would have been no question of bank failures or availability of customers’ money. There’s never a “run” on a storage facility where college kids store their furniture, for example, because that property is genuinely being warehoused.
LIFE INSURANCE DURING THE DEPRESSION
We have seen what happened to the banks during the early years of the Great Depression. What about the life insurance companies?
The number of life insurance companies did decline in these years, from 438 at the end of 1929 down to 375 at the end of 1933. Note that this decline of 14 percent was far lower than the 38 percent drop in the number of commercial banks during a comparable, but not identical, period.
Yet even this comparison may be too generous to the commercial banking sector, and too harsh to the insurance sector. Just because a particular life insurance company in, say, 1931 was taken over by a healthier competitor, doesn’t by itself tell us what happened to the customers of the failing insurer. I have seen (admittedly biased) insurance agents claim that not a single customer lost his or her assets as represented by whole life insurance contracts during the Great Depression, and thus far I have found no evidence to dispute these claims. (If any LMR readers have official sources either backing up the claim or refuting it, please let me know.)
To get a sense of the relative health of the insurance sector, we can quote from H&S:
On the surface, insurance companies were far from failure during the Depression. Official statements of the companies showed asset values comfortably in excess of policyholder reserves during the entire period. According to these documents, life insurance companies were in robust condition, even at the nadir of the Depression. At the end of 1932 the total assets of all U.S. life insurance companies were reported to be $20.7 billion, some $1.4 billion in excess of total liabilities, and $2.9 billion in excess of policyholder reserves [the present discounted value of expected future beneficiary payments— RPM]. Total capital of the insurance companies was reported to be $1.4 billion or 7 percent of total life insurance assets.
However, H&S go on to warn the reader that these official statements were potentially misleading, because the insurance companies weren’t valuing their portfolio of assets at prevailing market prices. For example, investment-grade bonds not in default were valued on their books at cost (adjusting for accrued amortization). In today’s parlance, the seemingly rosy report quoted above was not done with “mark-to-market” accounting.
Why would this matter? So long as the insurance companies were fine on a cashflow basis (and they generally were), what would it matter if the official market value of their assets temporarily dropped, due to the extraordinary financial crisis?
The potential problem was that a cashflow crunch could force the insurers to begin selling off their financial assets, in order to meet customer obligations. H&S provide some of the details:
“Insurance companies… stood in a perilous condition at the start of 1933. During the Depression policyholders markedly accelerated the rate at which they drew on the savings and credit features of their life insurance contracts. Cash surrender payments tripled, rising from $448 million in 1929 to $1.3 billion in 1932. As a result, insurance companies’ net cash flow dropped dramatically, from $1.5 billion in 1929 to $655 million in 1932. This limited the insurance companies’ ability to restructure their portfolios.
The dramatic increase in policy loans further restricted insurance companies’ portfolio choice. Total policy loans at all companies rose from $2.4 billion at the end of 1929 to $3.8 billion at the end of 1932. At the latter date they accounted for 18.3 percent of insurance companies’ reported assets.”
As Carlos and I explain more fully in our book, the issuers of whole life policies must invest the premium payments into various assets, to ensure their ability to pay the contractual amount upon death or maturity of the policies. The policyholder himself gets “first dibs” on these investable funds, in the form of a policy loan.
From the insurer’s viewpoint, policy loans are incredibly safe, because the underlying cash value on the policy serves as collateral. However, policy loans do have a downside, in that they are relatively illiquid (an insurer would have difficulty selling a given policy loan to another institution) and their payback schedule is uncertain. What H&S are emphasizing is that the pronounced increase in policy surrenders and policy loan requests in the early 1930s boxed the insurers into a corner, where they couldn’t invest their incoming premium payments the way they may have liked, given the rapidly changing economic landscape.
We come now to the punchline. H&S report that in addition to the well-known banking holiday, there was also an insurance holiday:
“The advent of banking holidays… further aggravated the situation of the insurance companies. With the banks closed or allowing withdrawals on only a restricted basis, people turned to their life insurance for cash. Like the banks, the insurance companies were faced with the possibility of a run that would force them into failure.
Rather than permit this to happen, the states took emergency measures. On 6 March 1933, the New York state legislature passed an act suspending the state’s insurance law and empowering Superintendent of Insurance, George S. van Schaiek “to make, rescind, alter and amend rules and regulations imposing any condition upon the conduct of any insurers which may be necessary or desirable to maintain sound methods of insurance and to safeguard the interests of policyholders, beneficiaries and the public generally,” during the emergency… The law took effect the following day and applied to all companies licensed to do business in New York state, not just those headquartered in the state.
Thus, the New York law covered most of the country’s insurance companies. In any case, it was soon copied by twenty-eight other states. The insurance holiday was under way.
On 9 March 1933, Superintendent van Sehaiek issued the first regulations following a meeting with representatives of the leading insurance companies. Effective immediately, insurance companies were prohibited from paying cash surrender values or granting policy loans in cash, although each policyholder could obtain up to $100 in the case of dire and demonstrated need… Moreover, policyholders could not withdraw any sums that they had left on deposit with the company. However, insurance companies were strictly enjoined to continue payment of death claims, annuities, and matured endowments…
The insurance holiday remained in effect long after banks had reopened their doors, although its terms were progressively liberalized. On 3 April 1933, the New York state regulations were amended to permit insurance companies to grant policy loans or pay cash surrender values for specified purposes such as the payment of rent or taxes where the insurance company “was satisfied that the applicant has no other reasonable means of meeting the necessity.” Policyholders were also permitted to withdraw all deposits made after 9 March 1933 and part of the deposits made prior to that date. On 7 June 1933, the New York state regulations were further amended to permit policyholders to obtain policy loans or cash surrender values upon stating in writing how they intended to use the proceeds. Thus, the insurance company no longer had to verify the policyholder’s need for the money… On 7 September 1933, van Sehaick declared the emergency over, and on 9 September 1933 all restrictions on policy loans and the payment of cash surrender values were removed six months after they were first imposed. The insurance law was back in force.” [Bold added.]
Although the insurance “holiday” is disconcerting, and shows that the contrast between banks and insurers during the Great Depression was not as stark as some may have thought, even so there is a very important difference that H&S don’t highlight in their own discussion.
It is crystal clear that the commercial banks were failing—by the thousands—and that customers were losing money, before the federal government stepped in to rescue the banks from their own precarious position in 1933. There was a total suspension of bank activity, meaning customers could not get their money at all. Furthermore—what H&S don’t bring up at all—even after the holiday was ended, the banks were still in a terrible position, as I explain in my book on the Great Depression (a story more elaborated in the sources I cite there).
In contrast, H&S haven’t really shown us that customers were hurt by the insurance companies. Their own figures and discussion show that going into 1933, the industry as a whole was still able to make its contractual payments (including policy loans), but that its ability to do so was being pushed to the edge. Even when the state governments intervened to provide relief, the insurers’ core business—providing death benefit payments to beneficiaries—was never interrupted. (Annuities and maturing contracts were also paid out in full, with no interruption at all.) Furthermore, policyholders were still able to get $100 in policy loans, so the suspension even here was not total.
Though the state-government- imposed insurance holiday no doubt was a burden to many people who wanted to obtain policy loans or to surrender their policies outright during the key months in 1933, it would be grossly inaccurate to conclude (as H&S seem to want to) that insurance customers suffered more than banking customers. All things considered, the conservative insurance sector weathered the Great Depression far better than the commercial banking sector. In fact, as happened after our most recent financial crash, many insurers saw a big increase in business in the immediate aftermath of the 1929 crash.
In future articles I will fill in the details of the fate of insurance policyholders during the Great Depression. My point in the present issue was to go through a hostile take on the matter—put out by authors sympathetic to the commercial banking sector—and show that their own analysis shows the relative superiority of insurance during our nation’s most terrible financial episode.
1. Thomas F. Huertas and Joan L. Silverman, “The Banking and Insurance Holidays of 1933,” available at: www.thebhc.org/publications/BEHprint/v013/ p0105-p0115.pdf.
2. Robert P. Murphy (2009), The Politically Incorrect Guide to the Great Depression and the New Deal, Regnery.