Economics 15: solvency and liquidity

by George Hatjoullis

The characterization of the relationship between the central bank and the banking system that was used in Economics 14 was general and was not the usual textbook exposition. The latter portrays the amount of credit the banking system can create as capped by the requirement for banks to hold reserves of fiat money in accounts at the CB. This is a rather simplistic exposition and one that does not apply to many important jurisdictions. The exposition used in Economics 14 is that bank money creation is influenced by the conditions under which banks can borrow fiat money from the CB and in particular the costs involved. The system is also constrained by the amount of capital committed to the banking business and regulatory capital requirements. However, banks can always raise capital, at a price. The two general elements are thus solvency and liquidity.

An entity is said to be solvent if the value of its assets exceeds or is equal to the value of its liabilities (the entity may be an individual or an institution). A liquid entity is one that can service or meet its ‘current’ financial commitments. A solvent entity however need not be liquid, nor, indeed, need a liquid entity be solvent. The two concepts play an important role in economics and yet are often conflated and misunderstood. This is particularly true in respect of banks and other financial institutions.

Solvency is in part a function of the valuation convention. If all assets are valued at cost (book cost) then the entity may be deemed solvent even if the current market value of the assets is less than the amount of liabilities. If the entity paid £100 for an asset and borrowed £100 to finance the purchase, then the entity is deemed solvent as long as the asset is valued at book cost. The entity may not be deemed solvent if the asset is valued at the price at which the asset could be sold today (marked to market) and this value is less than £100. If prices fluctuate daily the entity may be deemed solvent one day and then insolvent on another day. Evidently, solvency is a more slippery concept that it may first appear and one can see why valuation conventions might deviate from the intuitively appealing mark to market. The appropriate valuation basis for assets (and liabilities) depends upon the context and the economic role that the solvency test is intended to serve.

An entity may have continuous financial commitments. The entity is liquid if it has the wherewithal to fulfill these commitments. The loan will involve a rate of interest and possibly a repayment schedule and the entity will need money to meet the payments. The entity is deemed to be liquid if it can access the necessary money for payment. A business with a strong cash flow might have no trouble meeting ‘current’ liabilities even though it is, or is deemed, insolvent. This may reflect the valuation basis or the nature of the cash flow. It could be that the cash flow represents, in part, the continuous realization of the asset as cash. In this case there will be some debt left when the asset is fully realized. At this point solvency and liquidity converge. A solvent entity might be illiquid if it cannot generate sufficient money to meet continuous commitments. This may be because it cannot realize the assets (hence the concept of illiquid assets).

A bank has assets and liabilities on its balance sheet so the concept of solvency is clearly relevant. A bank also has cash inflows and financial commitments that necessitate cash outflows. A bank can be solvent but experience a liquidity crisis. The judgment on solvency is, in the last analysis, down to the bank regulator (which may be the central bank). If the central bank deems a deposit-taking bank to be solvent it will invariably help it overcome any liquidity problems. This is the lender-of-last-resort role of central banks. The terms under which the CB will provide such help will vary from jurisdiction to jurisdiction and may be deliberately expensive to ensure that banks manage themselves to minimize the need to resort to this facility. It is highly unlikely that any CB will allow a bank that it deems solvent to fail because of liquidity problems.

Banks hold capital (shareholder equity). This has been contributed by shareholders or accumulated through retained earnings. This capital can be used to acquire assets (loans). In addition, banks borrow from many sources to finance loans and other assets. Deposits are just loans to the bank. They leverage or gear these borrowings and, as we saw in Economics 14, create bank money. They simply lend by giving the borrower an account against which to spend and they can do this beyond the amounts that they have borrowed from deposits, shareholder equity or bonds. This is leverage. The question arises as to what limits or constrains the ability of deposit-taking banks to create bank money? Indeed how can the CB encourage them to lend?

As noted, Economics textbooks usually explain the limit in terms of the need to hold fractional reserves with the central bank. The required reserves limit the amount the banking system as whole can lend depending on the level of reserves required. However, this textbook explanation is oversimplified and the subject of quite a lot of controversy and debate. Moreover, it does not seem wholly consistent with the liquidity function of the CB. Surely, if a deposit-taking bank needs reserves at the CB it can borrow them, albeit at a price? Would the CB allow a solvent institution to run into a liquidity problem without aid? The answer must be unequivocally that it would not. Banking systems rely on confidence, and a CB would risk liquidity problems at one bank spreading to all if it did not assist any one solvent bank that has liquidity problems. Arguably this is the fundamental role of a central bank.

The precise details of central banking vary across jurisdictions. However, common to all are two essential and logical components, solvency and liquidity. The relationship between the CB and banking system can be best understood using these two concepts. The CB wants solvent banks in the terms defined by the regulator. The banks are thus constrained in their lending by their capital adequacy and the terms of solvency (although they can always raise more capital, at a price). However, subject to the solvency conditions being met, the CB in all jurisdictions acts to ensure banks do not run into liquidity problems, if only because this can undermine confidence in the banking system and result in a systemic crisis. The CB will define the conditions under which it will provide loans of fiat money to the banks and the cost of such loans. It may also require them to hold reserves in the form of deposits with the CB. However, it will also stand willing to lend to fund such reserves. Reserve requirements are thus not a physical constraint on bank lending though the cost of borrowing and holding reserves may well discourage such banks from relying on this source. Additional constraints may arise through the conditions for borrowing from the CB, notably eligible collateral.

This brings us full circle in the discussion on solvency and liquidity. A solvent entity can potentially resolve a liquidity problem by borrowing against illiquid assets. The assets are pledged as collateral in case the loan is not repaid. Borrowing from the CB is normally against collateral but not all assets are deemed eligible as collateral. An individual bank in need of CB liquidity might have to incur costs to acquire necessary collateral and is thus likely to ensure it holds such a ‘sufficient’ quantity of such assets as part of its balance sheet. Liquidity is normally available but at a price. Individual, well capitalized, banks might constrain their lending because they feel that if they need to access CB loans, the limited supply of available eligible capital might make this exercise prohibitively expensive. The result might be that the system as a whole provides less credit than the capital adequacy allows and that the central bank would like. In such circumstances the CB could relax the collateral eligibility rules but it need not.

The important point to take from all this is that bank money creation is influenced through two channels. First, there is the capital adequacy and solvency test that each bank, and hence the system, must pass. Second, there are the conditions under which the CB will provide fiat money to the banking system. The assertion made here is that the CB will never deny fiat money to a solvent bank experiencing liquidity problems. To do so would court systemic problems and somewhat contradicts the idea of the CB as lender of last resort. The fiat money, however, will come at a price and in taking account of this price banks will take the risk of illiquidity into account in their lending practices.

The maturity of borrowing from the CB is typically very short. The CB is not there to fund bank activity. It is there to ensure that the liquidity of the assets of solvent banks does not become a problem. Control is exercised through the cost of borrowing from the CB and the unpredictability of such cost. It is not a good idea to make loans at 3% and to fund such assets at more than 3%. In general the CB should be seen as controlling the creation of bank money via setting the price (the interest rate) and then making available whatever fiat money the system demands at this price. If it wants less bank money creation it raises this price. It can lace this basic framework with physical controls but the main instrument of policy is the price. The interest rate also has the virtue of being symmetric in encouraging and discouraging bank money creation in equal measure. Controls are asymmetric. They may succeed in physically restricting bank money creation but they cannot be used to force banks to lend.

The purpose of the Economics series is to provide sufficient conceptual tools to enable the uninitiated to grasp the economic issues of interest in their jurisdictions. The series is thus necessarily lacking in specific detail of institutional arrangements and practices in any one jurisdiction. In part this is because the author is unaware of these details! To write from a UK perspective would be unhelpful. The result is a rather general but hopefully more accessible and relevant exposition than offered by the typical economics textbook.

postscript 18/03/2014

Bank of England have published a very good article on the process of money creation.