If I said that the recent recession was caused by a shortage of liquidity, most people would say I was mad. "Look how easy it is to get a loan, at such low interest rates!"
In hyperinflations, the price of liquidity is extremely high. If the price level is doubling daily, then currency pays a real interest rate of minus 50% per day. That is an extremely high price to pay for holding a liquid asset like currency compared to holding an illiquid real asset like rice. And that extremely high price of liquidity in a hyperinflation will make trade more costly, and will reduce the volume of trade. The volume of trade also falls in a recession, but recessions aren't like hyperinflations.
"MONEY is not, properly speaking, one of the subjects of commerce; but only the instrument which men have agreed upon to facilitate the exchange of one commodity for another. It is none of the wheels of trade: It is the oil which renders the motion of the wheels more smooth and easy." David Hume.
Let's take Hume's metaphor semi-literally, and see how far it takes us.
Suppose we need trucks to trade goods, and trucks need (diesel) oil. If a fall in the supply of oil causes a rise in the price of oil, trade becomes costlier, and the volume of trade falls. Something like that probably happens in a hyperinflation. (It probably happens in moderate inflations too, but maybe the effects aren't big enough to see clearly.)
But when we say there's a "shortage" of (diesel) oil we might mean two different things: we might mean the supply is low and the price is high; or we might mean that gas stations have run out so truckers aren't always able to fill their tanks. A shortage of oil in that second sense would also disrupt trade; some trucks would run out of oil and be unable to transport goods; and some truckers with full tanks might refuse to deliver goods for fear of being unable to refill their tanks.
A hyperinflation is more like a high price of oil; a recession is more like a high risk of not being able to buy oil.
But that is probably about as far as we can push David Hume's metaphor.
In a recession the unemployment rate increases. People find it harder than normal to sell their labour; they find it harder to buy money with their labour. And firms find it harder than normal to sell goods; they find it harder than normal to buy money with their goods. That looks like a shortage of money to me, in that second sense of "shortage".
But if you have sufficient safe colateral, you will have no difficulty in borrowing money in a recession. The only question is: at what price?
1. If everybody knew that the recession would be temporary, that would presumably reduce the fraction of income people wanted to save, which would imply a higher equilibrium interest rate on safe loans. And if this were the only factor influencing interest rates we would expect to see recessions associated with higher (real) interest rates. (This is what would be predicted by simple macro models if recessions were caused by the central bank causing the money supply to fluctuate.)
2. But recessions also make some previously safe assets riskier. For example labour becomes a riskier asset if the risk of unemployment increases. The pool of safe potential borrowers, and the amount each one could safely borrow, shrinks in a recession. And borrowing to invest to produce extra goods looks less attractive, if there is a greater risk you will be unable to sell those extra goods in future. These factors mean we would expect to see recessions associated with lower (real) interest rates on safe loans.
In a representative agent model, where unemployment is shared equally (so everyone is on short hours), and with no colateral constraints, and with no investment, the first of the above two factors would predominate. The representative agent would want to borrow in a temporary recession to smooth consumption, so you could only get a recession if the central bank set the interest rate above the natural rate.
Throw heterogeneity, risk of default through unemployment, investment, and the perception of a longer-lasting recession, into the model, and it might well be the case that the second of the above two factors would predominate. The safe interest rate might be below the natural rate in a recession. The IS curve might slope the wrong way.
There is no contradiction between a liquidity "shortage" and low nominal interest rates. (I hear that Venezuala has very cheap gas, if you can get it.) And because people can't be sure they can easily buy money, they will tend to want to hoard more than they usually would, and be less willing to sell it. Which makes it harder for other people to buy money.