How much of a problem in the theoretical/practical world is the problem of illiquidity? For instance suppose you own a corporate bond/OTC contracts with risk based in Asia and which are relatively illiquid. Then suppose latin America market bottoms out resulting in the current tradeable price of your instruments to fall . (Risk in Latin America is not linked to Asia but I’m going along the lines that investors do not behave rationally in a collapse and tarnish all instrument with the same brush. Collapse in Latin America does not lead to collapse in Asia but makes investors reevaluate the risk associated with holding assets in these these emerging countries). My point is if this happens does the fact that I cant get rid of the asset (risk based in Asia) a problem for me. (Would it affect capital requirement in some way etc)?

Illiquidity is a big problem if you are a forced seller (or buyer for that matter) - if you are in for the long haul then it is less of a concern.LTCM's problems were largely blamed on a lack of liquidity, although it seems to me that lack of liquidity is something people often blame when the price moves against them - "there weren't enough buyers for the amount I needed to sell, the price went down, its not my fault its the markets lack of liquidity"..On the assumption that your Asia asset now needs to be marked at a lower market level, reducing the overall value of your assets, I imagine that capital requirements would be affected, although I'm not an expert on how. I guess it will depend on the regulatory regime under which you operate.

It's hard to answer this question in general.Liquidity risk is modeled separately than market risk. It's possible to have a positive net present value and still be unable to meet your financial obligations. Liquidity risk charges are a significant fraction of total risk charges for many organizations.Illiquidity is a major difficulty in modeling. For one thing, it means price data are less reliable. For another, it makes a big difference for many models whether you have jumps, when you cannot transact at prices between S(t) and S(t+e), and extremely high volatility, when |S(t+e) - S(t)|/S(t) is a big number but you can transact at intermediate prices.But you are talking about something different, like someone saying they have a "cash flow problem" when they are really insolvent. You are saying you own Asian securities when the Latin American market crashes. Your securities "shouldn't" decline in price, but they do. You don't want to realize the loss, so you say it's a "liquidity problem" and switch from mark-to-market to matrix pricing (mark-to-model). This happens a lot, but it's not a liquidity issue.