Consolidating Central Bank and Central Government Accounts – The Mark to Market Problem
One of MMTs main principles is that (put simply) in a sovereign currency the central bank and state can be treated in accounting terms as a single consolidated entry. Neil Wilson makes a brilliant case for this in his blog. But although this produces important insights I am skeptical, like a number of others in the post-Keynesian community, whether this is entirely accurate in terms of modelling central bank operations.
I am not going to be dogmatic about this, I am open to being convinced. I am not a professional accounting but I do see a problem in accounting terms. This post is basically a call for opinions and expertise. The heterodox community, by its very nature, requires this. The problem, broadly, is this:
In consolidated accounting you basically a subsidiary company owned (100% equity) by another body as if it doesn’t exist. All of the subsidiary body’s assets and liabilities appear on the parent body’s balance sheet, and all of the subsidiary body’ revenue, expenses, gains and losses appear on the parent company’s income statement. The body’s financial results, therefore, are consolidated on a single set of statements. In such an arrangement any loan and between subsidiary bodies will cancel out as an asset liability pair, as will profits and expenses from intragroup sales. To do otherwise would be double counting.
However there are is a special issue that arise and have particular relevance when the subsidiary body is a bank. These issues relate to the economic value of the assets of the bank and related to this the treatment of goodwill. Both if these ultimately resolve into the valuation of the ‘fiscal backstop’ the state provides to a central bank, and indeed all banks.
When a holding group acquires a subsidiary then it may do so at a value that is greater than the value of the acquired company alone. That difference is marked down on balance sheets as goodwill. In the UK as I understand it that goodwill is amortized over time, in the US in recent years (and controversially) it is continuously revalued at mark to market (fair value). Although the acquisition of a bank may have occurred many eons ago it is appropriate to continuously revalue goodwill as the additional value that the Central Bank gains as being wholly owned by the state. What is this additional value?
There are two aspects to this. Firstly a bank, any bank, has an intangible asset (goodwill) – called in the finance community the ‘charter value’ of the bank. If you simply swept up the accounts (as you do when a firm closes) netting out assets and liabilities the equity holders get the difference. In this process you lose the goodwill of the firm, the value of the additional profits that the firm earns from being a firm – the market power/power of the brand/IP of that firm (which incidentally explains the profit puzzle). In short it is the value of the future profits the firm may make over and above the value of its tangible assets. Banks in particular may have an ‘economic value’ in the long term which is well above the ‘mark to market’ value of their assets. This is excellently highlighted in posts by Stephen Randy Waldman and John Hempton on the current solvency problems of banks, and the need for the state to step in or order to realise any long term economic value that banks have.
In effect the state provides this through the ‘fiscal backstop’ that is the implied guarantee that the state gives to a Central Banks that if the Central Banks has a negative balance sheet it will step in. So for example when the Fed propped up through open market operations various failed investment banks the Treasury wrote letters guaranteeing that any balance sheet losses of the operation would be born by a diminution of the normal profits the Fed pays to the Treasury. Why should a state rationally do this (setting aside for a moment whether nationalistation would have been a better long term deal), because in the long term the charter value of the saved bank would (in theory) give it a positive economic value not reflected by its mark to market value in a financial crisis. The fiscal backstop is equivalent to unlimited liability, an investor increasing its equity holding without increasing its equity share. (of course one of the big problems with teh ECB is who provides the backstop – hence Draghi has had to invent one).
In the past I have championed the ‘charter value’ theory of banking promoted by Neil Wilson and then taken up by Professor Steve Keen (see here). In this a bank has a ‘charter value’ an intangible asset resulting from its licence to trade as a bank. I originally sought to correct this by applying the fundamental equation of accounting, as under this theory starting a bank would be impossible, you would need initial owners equity, so I modified it so the charter value of a start up bank was equal to the original equity, I now see my original modification was not far reaching enough, rather the ‘lending power’ (charter value) of a bank has two components – a tangible asset created by equity and returning to equity, and an intangible asset which does not return to equity on bank closure. The capitalization of both is the economic value of the bank from lending as a going concern. The first element results on profits from lending from prior savings (converted to equity), but any firm can do this. The second is unique to a bank (or shadow bank) and that is ‘maturity transformation’ or as I like to call it ‘liquidity transformation’. That is the ability to create money ‘out of thin air’ because the depletion of reserves (creation of liabilities) from the loan will be met in a very short time from interest and principal payments on other loans and net deposits from economic growth and the increased desire to hold idle balances from wealth effects (the reflux effect). Banks in effect ‘borrow’ idles reserves (liabilities) and treat them for a time as assets providing they can stay liquid and pay back the liabilities – the fractional reserve model then Venice banks discovered in the C14, they could make money from the money idle in their vaults. Today the central bank reserve window acts as a backstop in cases of temporary lack of liquidity. This reserve window only works however because of the assumed mid-long term solvency of the lenders (hence the Bagehot rules). If the banking system is insolvent that either their is widespread bank failure of the Central Bank, with its fiscal backstop, props up the system with the expectation that this investment will be rewarded with future growth and tax revenues.
The simple point I am making in this post is that you can only correctly model the central bank/treasury relationship after modelling a) the charter value of the bank and b) the implied equity level of the fiscal backstop. Otherwise you will be missing the full liabilities the treasury. Once you have modeled this bilateral (dialectical if you like) relationship then and only then can you consolidate them. But you can never assume you can miss a step and then leap to consolidation – that is making an accounting mistake. So then I submit there is nothing lost, and much gained, as Lavoie has stressed, from modelling the government as a the outcomes of the whole central bank – treasury -spending – taxing system, rather than solely as a black box system which pumps money out through spending and nets its out through taxes.