Banks and building societies operating in the UK suffered an aggregate net loss after tax and provisions of £5.8 billion in the first three quarters of 2008, according to data obtained from the Bank of England under a freedom of information request. This compares with a net profit for banks alone of £29.0 billion in all of 2007. (The statistics include building societies from the start of 2008.)

The loss over Q1-Q3 2008 was due to a negative contribution of £18.9 billion from dealing activities – probably reflecting write-downs on securitised assets – and provisions of £12.8 billion for bad and doubtful debts. Pre-tax profits excluding dealing and provisions were £26.6 billion versus £40.5 billion for banks alone in all of 2007 – see chart.

Despite the £5.8 billion loss, banks and building societies paid out dividends of £17.9 billion over Q1-Q3 2008. Consequently, their stock of retained earnings fell by £23.6 billion – more than cumulative retentions by banks alone over the previous three years.

Even allowing for a further retained earnings loss in the fourth quarter, the erosion of internal capital last year will have been comfortably exceeded by new capital-raising of more than £70 billion, with the government contributing £37 billion. In other words, despite staggering write-downs and provisions, banks’ and building societies’ aggregate capital ended 2008 significantly higher than a year before.

November trade figures released today are grim and show little evidence of the benefits promised by the many advocates of sterling depreciation, including Bank of England Governor Mervyn King.

The ratio of export volumes to import volumes, excluding oil and erratic items, slumped to its lowest level since December 2007 – see first chart.

Trade adjustment takes time but manufacturers are likely to have been constrained from taking advantage of improved price competitiveness by their lack of access to credit. As previously argued, sterling’s plunge may have contributed to credit restriction by accelerating a withdrawal of foreign funds from the banking system.

Meanwhile, the annual increase in manufactured import prices reached 14% in November and may hit 15-20% soon as a result of the further fall in the exchange rate – second chart. Manufactured imports account for 17% of domestic demand – the 14% rise implies a cut of 2.4% in real domestic purchasing power.

Devaluationists will no doubt argue that trade performance would have been even worse but for the exchange rate fall, while conveniently ignoring its negative effects on spending power and credit availability. The myth that the recovery of the early 1990s depended on a prior collapse in sterling continues to exert a strong influence.

Insurance companies and pension funds increased their liquid assets – currency, bank deposits and short-term money market paper – by £28 billion, or 18%, in the year to September. With the value of their portfolios falling by 10% over the same period, the ratio of liquid to total assets rose to 9.1% – the highest since September 1990.

Even assuming no further addition to liquid assets in the fourth quarter, weakness in markets should have ensured a continued rise in the ratio. In other words, the liquidity ratio has probably now surpassed the 1990 peak and is at its highest level since 1974 – see first chart.

Institutions generally target a stable proportion of liquid assets in portfolios over the medium term. For insurance companies and pension funds in aggregate, the ratio has averaged 5.5% since 1964 and the chart shows evidence of mean reversion.

Suppose insurers and pension funds attempted to reduce the liquidity ratio to its long-run average of 5.5%. Based on the position in September, this would involve a first-round injection of £70 billion into markets. Institutions might take the opportunity to "rebalance" their portfolios towards assets that have underperformed recently, including UK equities and corporate bonds. (Overseas investments have benefited from the fall in sterling.)

The £70 billion of buying, however, would represent only the first stage of a multiplier process. An individual institution can reduce its liquidity ratio by buying assets but if the seller is also an insurer or pension fund the aggregate position is unchanged. Rather than being extinguished, "excess" liquidity is simply transferred to another institution, leading to a further round of buying.

Consider the extreme case of a fixed supply of assets entirely owned by insurance companies and pension funds. Liquidity would circulate between institutions, stimulating further buying, until asset prices rose sufficiently to bring the aggregate liquidity ratio down to its target value. In other words, asset prices would bear the entire burden of adjustment back to equilibrium.

In practice, the supply of assets available to insurers and pension funds is not fixed – higher asset prices will induce other holders (e.g. overseas investors) to sell and originators to issue more. However, the essential point remains – the attempt to restore liquidity equilibrium is likely to involve multiple rounds of buying and a significant impact on asset prices.

Consistent with this explanation, there is historical evidence of a positive relationship between the level of insurance companies and pension funds' aggregate liquidity ratio and future UK inflation-adjusted equity returns – second chart. The line of best fit suggests each 1 percentage point rise in the ratio is associated with a 10% increase in the cumulative real return on equities over the subsequent five years.

One caveat to the above analysis is that institutions may wish to hold a higher proportion of short-term assets in their portfolios than in the past because they have entered into interest rate swap agreements involving payment of a floating rate in exchange for a fixed nominal or real income stream, intended to match future liabilities. This may have raised the "equilibrium" level of the liquidity ratio, although any increase is likely to have been much smaller than the recent actual rise, implying the latter still has positive implications for asset prices.