Editor’s note: The following is Cairn Cross’ written testimony submitted to the Legislature on Jan. 22. Cross is a former banker and a founder of the venture capital firm FreshTracks Capital.
The Bank of North Dakota is owned and operated by the State of North Dakota under the supervision of the Industrial Commission as provided by Chapter 6-09 of the North Dakota Century Code. The Bank of North Dakota combines elements of banking, fiduciary, investment management services, and other financial services, and state government and plays a primary role in financing economic development.
The Bank of North Dakota is a participation lender; the vast majority of its loans are made in tandem with financial institutions throughout the State of North Dakota. Its primary deposit products are interest-bearing accounts for state and political subdivisions. By statute all North Dakota State funds and also funds of the state penal, education and industrial institutions must be deposited in the bank and cannot be deposited in commercial banks.
The Bank of North Dakota will accept deposits from retail and commercial customer and other institutions. However, it has only one branch and makes no attempt to compete with the retail banking arms of North Dakota banks. I have heard from Vermonters that municipalities are required to deposit their municipal funds with The Bank of North Dakota. I have not been able to verify that by reviewing the audited financial statements of the bank.
The loans the Bank of North Dakota makes can be segmented into four types: student, commercial, agricultural and residential mortgages. One way to think of the Bank of North Dakota in relation to Vermont is that it is similar to combining the “state component units” the Vermont Student Assistance Corporation, the Vermont Economic Development Authority (commercial/agricultural), and Vermont Housing Finance Agency (residential mortgage) into one centralized financial institution. However, where VSAC, VEDA and VHFA rely on private activity bonds and commercial paper to finance their lending activities, the Bank of North Dakota relies on deposits mandated by statute.
North Dakota requires that the Bank of North Dakota be the intermediary of state funds, where Vermont permits commercial banks to intermediate state funds through their normal banking activities. Vermont funds VSAC, VHFA and VEDA loan activity through the issuance of bonds, notes and commercial paper. One could argue that the deposits used by the Bank of North Dakota provide a more stable and perhaps cheaper way to finance loan activity over the long run. VSAC, for instance, has had a less than stable ride from the bond markets during the past two years
At year-end 2008, the Bank of North Dakota’s loan portfolio consisted of approximately $776MM in student loans, $1.065BN in commercial loans, $509MM in residential mortgages, and $268MM in agricultural loans for a total loan portfolio of $2.618BN. These loans were supported on the balance sheet by $2.645BN in deposits for a loan to deposit ratio of close to 100% (98.9%). For comparison purposes BISHCA’s Consolidated Comparative Statement of Condition and Analysis of the 15 Vermont and National Financial Institutions (primarily commercial banks) domiciled in Vermont reveals that the consolidated loan-to-deposit ratio of these Vermont banks was approximately 93% at December 31, 2008. It’s important to note that there are other banks operating in Vermont that are not part of BISHCA’s report (Citizens Bank, Key Bank, TD Banknorth) and the loan-to-deposit ratios of these banks might differ significantly and the sheer scale of their banking activities might skew the state’s bank loan-to-deposit ratio. Still, the
Bank of North Dakota appears to use a greater percentage of its deposits to fund loans than does the Vermont banking system. Perhaps this has to do with the Bank of North Dakota’s economic development mission.
It is interesting to compare a few key metrics for the Bank of North Dakota to a theoretical consolidated VSAC, VEDA and VHFA. I note that on June 30, 2009 VSAC had a student loan portfolio of $2.29BN (almost three times the size of BND’s student loan portfolio) and had borrowed $2.309BN from bondholders to finance those loans. On June 30th 2009, VEDA had loans receivable (both Agricultural and Commercial) of approximately $102MM (about 8% of BND’s total commercial and agricultural loans), which were financed by approximately $97MM of notes and commercial paper. VHFA had a residential mortgage loan portfolio of approximately $681MM which was financed by borrowing $739MM from bondholders. It’s important to note that VEDA originates a number of loans on behalf of other programs and the loans do not remain on VEDA’s books. For instance VEDA originates some tax exempt bonds for college buildings. They originate loans for Vermont’s Clean Energy Development Fund, too.
It’s possible to conceive of consolidating VSAC, VEDA and VHFA into one new state-owned bank. However, there would be some large challenges. For instance VSAC and VHFA are arguably over-financed, having borrowed more from bondholders than the value of their loan portfolio.
In fairness, each of these entities has other restricted and unrestricted assets which could be absorbed into a consolidated entity. But some of these assets (leasehold improvements for instance) might have limited value in a consolidated entity. In order to assume the VSAC and VHFA portfolios, the new state bank would need to raise more than $3BN in deposits, or it would need to assume the responsibility for the current bond financing.
It might be possible to sell a considerable portion of VSAC’s loan portfolio, repay debt and thereby reduce the amount of deposits needed to start a new state owned bank. For instance one might consider selling all student loans made to non Vermonters and prohibiting the origination of new student loans to non Vermonters. My assumption is that lending to out of state students is one of the ways BND and VSAC differ.
Furthermore, it would likely not be easy to convince Vermont banks that they should give up their governmental deposit business. Finally it’s important to note that the Bank of North Dakota is not an FDIC insured bank. All deposits are guaranteed by the state of North Dakota’s treasury.
Presently in Vermont the bondholders of VSAC, VEDA and VHFA bear the risk of loss with at most a “moral obligation” of the state to step in if one of the entities became insolvent and could not make its bond payments. However, one might argue that this moral obligation is really no different than an express guarantee of state bank deposits. Would the state truly allow VSAC, VEDA or VHFA to fail? If the answer is “no,” then why not guarantee the deposits in a state bank? It is essentially the same thing.
What might the rewards be of Vermont owning and operating its own bank? An examination of the Bank of North Dakota’s audited financial statements reveals that it transferred $30MM to the North Dakota General Fund in 2006, 2007 and 2008.
It has a long history of paying dividends to the state. During at least one year early in the past decade it acted as a “rainy day fund” for the state and transferred a “special dividend” to help the state of North Dakota balance its books.
A review of VSAC’s audited financial statements reveals that Vermont appropriated $20MM for VSAC in FY 2008 and $19MM in FY 2009. A review of VEDA’s financial statements reveals that VEDA transferred approximately $475K to the state of Vermont in FY 2008 and received a $222K appropriation (providing a net inflow to the state general fund).
During FY 2009 the state transferred or appropriated approximately $1.9MM to VEDA. The delta between the dividend received by the state of North Dakota from the Bank of North Dakota ($30MM) and the dollars paid by the State of Vermont to the “V” entities (approximately $21MM) is intriguing to note.
To be fair if, Vermont required state deposits to be placed in a state-owned bank, Vermont would lose the franchise tax on the deposits that Vermont banks currently pay. The Franchise tax is .000096 of the average monthly deposits. So if Vermont needed a theoretical $3BN in deposits to fund a state bank, and it disintermediated those deposits from Vermont commercial banks, it would lose approximately $3.5MM in franchise tax revenues.
One of the more interesting things to note is that in 2008 the Bank of North Dakota paid salaries and benefits to its employees of $9.5MM. During FY 2009, VSAC’s salaries and benefits alone were $25MM. During FY 2009 VEDA’s salaries and benefits were $2.1MM. During FY 2009 VHFA’s salaries and benefits were $3.3MM. In terms of efficiency, the Bank of North Dakota spent $9.5MM in FY 2008 to originate and/or service $2.6BN in loans.
Expressed as a percentage it cost the Bank of North Dakota, 3 tenths of one percent of their loan assets from a salary and benefits perspective to originate and service their loans.
During FY 2009, VSAC, VEDA and VHFA originated and/or serviced a combined $3.1BN (and perhaps another $75MM of loans originated by VEDA but not kept on its books. During FY 2009 these same three entities spent a combined $30.4MM to originate and service their loans. Expressed as a percentage the salary and benefit cost of the combined “V” entities was almost 1% of their loan portfolio.
One might argue that this lack of efficiency in the V entities is because VSACs portfolio is made up of a myriad of relatively small loans (more servicing and origination costs per loan) while the Bank of North Dakota’s portfolio is skewed towards larger business and agricultural loans (less servicing and origination costs per loan).
It could also be argued that VSAC’s grant-making and counseling functions are more robust than the Bank of North Dakota’s and take more people to operate. A counter argument is that BND also operates a deposit gathering and processing function (check clearing, processing and reconciliation) which is a labor intensive activity and manages to do all of this and make loans at a far more efficient cost.
One could theorize that there are other cost savings that might be achieved by consolidating the “V” entities into one bank. For instance, each V entity has a separate audit. Each V entity has overlapping and in some cases duplicative senior management teams. Each V entity operates its own IT and management information systems. It would not be a stretch to believe that perhaps $5MM to $10MM of expenses could be eliminated through consolidating the “V” entities.
It’s important to note that the Bank of North Dakota provides various other financial services to North Dakota banks and businesses. The Bank of North Dakota operates an interesting “start up business guarantee program” whereby the bank guarantees up to 85% of $100,000 of bank borrowings incurred by a beginning entrepreneur.
By statute the bank provides various credit facilities to states and municipalities, such as borrowings for telecommunications infrastructure and water districts, developmental disability infrastructure and “centers of excellence.” The Bank of North Dakota provides basic correspondent banking services such as check clearing to North Dakota’s commercial banks and credit unions. It will buy trust-preferred securities to provide a layer of capital to North Dakota banks. It has a lending arm that will lend to employee-owned companies. It has in fact lent money to commercial bank employees to buy their employer bank from its former owners. It provides escrow and trust services to the state and municipalities. It operates a small venture capital fund in partnership with the North Dakota State Economic Development department.
Perhaps it makes sense to consider other “V” entities that might be consolidated into a new state-owned bank. For instance entities where periodic state appropriations have been made for certain financial related services could be consolidated into a state bank. These entities might include the Vermont Sustainable Job Fund and the Vermont Employee Ownership Center and the Vermont Entrepreneur’s Seed Fund. It probably makes sense to consider consolidating some of the functions of the Vermont Pension Investment Committee into a state bank. Vermont Energy Investment Corporation often contracts with banks to provide certain lending functions, particularly for energy efficiency improvements for commercial and retail customers. It may be possible to consolidate parts of this entity into a state bank. The Vermont Municipal Bond Bank would be another strong possibility for consolidation. Finally the VEPC/VEGI programs seem like a natural to be operated by a state-owned bank.
There are many nuances of the Bank of North Dakota and the various V entities that have not been properly explored and analyzed in this report. But it is clear that creating an efficient newly chartered state bank will be a time consuming project involving considerable study and skilled execution.
At first blush though, the potential efficiencies and the streamlined governance seem worthy of consideration. If there is the political capital and will to “rethink” government this year, it seems as though consolidating many independent component units into one streamlined entity and creating a method to fund this entity with a stable reasonably priced capital in the form of deposits should be given proper attention and study. It appears at least theoretically possible to create an entity that operates far less expensively and at a profit and therefore is able to pay a dividend to the state each year rather than the reverse whereby the state needs to appropriate funds for these organizations on a regular ongoing basis.