Asset market liquidity varies considerably over time a fact that became painfully clear during the recent financial tsunami. "Illiquidity" can be defined as the price impact of buying or selling a given quantity of assets. "Liquidity" is the opposite of illiquidity and fell precipitously during 2008.

Until recently, there have been limitations to empirically testing why liquidity varies at the market level. Mark S. Seasholes of HKUST and his co-authors address this gap by studying the positions and trading revenues of New York Stock Exchange (NYSE) market makers. On the NYSE, market makers are also called "specialists". Seasholes and co-authors use proprietary data to calculate specialists aggregate inventories (positions) and specialists aggregate trading revenues.

Specialists inventories and trading revenues can predict future changes in the markets liquidity. When specialists hold large, risky positions today, liquidity falls in the coming days. When specialists have lost money recently, liquidity also falls in the coming days. The drop in liquidity is measured by larger bid-ask spreadsi.e., larger costs of buying or selling a set quantity of stocks. The authors write: "When specialists find themselves with larger positions or lose money on their inventories, the effective spreads are significantly wider in the days that follow."

The ability of inventories and trading revenues to predict future liquidity is not linear. The authors find that "when inventories are particularly large in either direction, market makers may be more constrained and require more compensation to provide liquidity. The data reveal strong evidence of this non-linearity. On typical days, each additional $100 million in aggregate inventories implies a next-day, market-wide effective spread [widen]. But when inventories are beyond the 75th percentile, the sensitivity of spreads to inventories more than doubles."

To summarize the papers first result: financial intermediaries can affect asset prices and asset liquidities. An otherwise healthy market, with financially healthy participants, can suffer price and liquidity drops due to intermediaries. The recent financial tsunami provided an example of just such an effect. Financial intermediaries (banks) lost money on investments such as mortgages and CDOs. The loses made the banks cutback on their day-to-day activities. As loans from banks dried up and trading became sparse, values of assets originally unrelated to mortgages began to fall.

The authors also classify specialist firms into sub-groups in order to better understand the role of intermediaries. During the sample period there were a substantial number of mergers and only seven specialist firms were left by the end of the study. In addition to the mergers, specialist firms have different ownership structures. Some firms are corporately-owned and some are partnerships. The authors hypothesize that a corporate ownership structure provides a substantial "financial cushion" which helps ameliorate the effect specialists have on the markets liquidity.

"The sensitivity of liquidity to inventories and revenues is greater for specialist-owned firms compared to corporate-owned ones. This is reduced when specialist-owned firms merge with corporate-owned ones, consistent with deep pockets easing financing constraints," they say.

"Our results should generalise to other market structures. Specialists have become much less important [on the NYSE] since 2007, but other liquidity suppliers, such as quantitative hedge funds, have taken their place and face the same kinds of financing constraints. In fact, given the US credit crunch of 2007-08, the financing constraints are probably quite severe at the moment."