Tuesday, 24 September 2013

Collateral Shortages and the Availability of Credit

Peter Stella has an interesting post on VOX on the implications of collateral supply for different approaches to running down the large balances of reserves held in the banking system as a result of QE operations.

Stella has written a number of papers and articles on the workings of the repo market often looking at the impact of collateral shortages on overall lending volume.   Together with Manmohan Singh, he has probably done the most to raise the issue of collateral availability and its implications.

In this most recent article, Stella compares two alternatives for managing a drain of large amounts of excess reserves.  The first is to simply offer banks term deposit facilities with the central bank.  This would effectively convert their holdings of short term claims on the central bank into a longer term one.  The second is a central bank reverse repo programme.  This would involve the central bank selling assets from its portfolio with an agreement to repurchase at a fixed price after a period of time.

Stella is in favour of the latter and believes that it will facilitate more credit creation.  He has two reasons for thinking this.

The first reason relates to banks balance sheets and the leverage ratio.  To understand this point, it is useful to note that non-banks cannot hold reserves.  Therefore, when the central bank acquires asset from non-banks, this necessarily involves adding both an asset and a liability to the balance sheet of the banking sector.  Reserves (assets of the banks) go up and deposits (liabilities of the banks) go up.

Although this may involve little risk to the bank, it is not without consequence.  Whilst holdings of reserves do not attract a capital requirement, they are included in the measurement of assets for the leverage ratio.  Where banks are constrained not by capital, but by overall leverage, holdings of reserves may be effectively crowding out lending to the private sector.

Switching reserves into term deposits will do nothing to change this.  However, if the central bank conducts reverse repo with the non-bank private sector, that will simultaneously drain reserves and reduce deposits, thereby eliminating the middle-man role required of banks.  For banks subject to a leverage ratio constraint, that may free up lending capacity.

It is hard to tell how important this is.  Certainly some banks have indicated that they expect to be impacted by the leverage ratio, over and above the normal capital requirements.  But how much difference that will make in the long run is another matter.  My own feeling is that is there would be some effect, but it may not be that great.

It is worth noting however that this aspect has nothing do with collateral.  The important distinction here is between arrangements transacted with banks and those transacted with non-banks.  If the central bank was to offer term deposit facilities to non-banks, this would achieve the same reduction in bank balance sheets as a reverse repo programme.

The second argument concerns the role of collateral chains in credit creation.  Stella points out that increased bank reserves do not in fact facilitate greater bank lending, as is suggested by the money multiplier model.  However, high grade collateral, according to Stella is crucial to the volume of lending within the non-bank sector, because of the amount of this that is carried out through repo.  Collateral shortages squeeze this form of lending.

I have written about collateral availability and the impact on loan volume before (here).  As discussed in that post, I do think there are good reasons why we might expect to see a positive correlation between the level of repo business and the amount of eligible collateral available to the market. 

However, the point I wish to make here is that, in almost every case, the credit creation associated with collateral availability is ultimately only financing the actual holding of that collateral.  It is not going to finance new expenditure on produced goods.

If we are concerned about low levels of lending, the sort of thing we are probably focused on is a small businesses that wants to incur investment expenditure but cannot raise the funds.  We want to see more bank lending so that this type of expenditure can take place.

Making more collateral available to the market does nothing to facilitate this.  The small businessman cannot use the extra collateral available to help raise the funds he needs.  Banks might be prepared to provide him with repo finance, but to take advantage of that he has to get hold of that collateral in the first place.  He can only do that by using the very funds he raises to buy or repo in the collateral.  The only people who can benefit from repo funding are those who are looking to take a position in the underlying collateral or others in the collateral chain.

This is not to say that this type of finance has no macroeconomic effect.  The use of repo serves to increase demand for the underlying collateral, which pushes up asset prices generally (see again my earlier post).  However, it is important to understand that is only through this effect on asset prices that any potential macroeconomic benefit is arising - there is no separate credit creation for the real economy that is being facilitated.  It therefore needs to be assessed in the context

Other things being equal, more liquidity is a good thing.  I'd therefore be inclined to take the view implied by Stella that draining reserves through a reverse repo programme available to non-banks is better than a bank only term deposit facility.  However, I think it would be a mistake to confuse any impact on the repo market with a potential improvement in finance availability for the real economy.

Tuesday, 17 September 2013

On the Role of Models

In the wake of Paul Krugman's post on Wynne Godley and hydraulic modelling, I've had a number of discussions, including this one with Phil Pilkington, on the role of modelling in economics.

If you've read a few of my posts, you will know that I use models a lot.  This in part a reflection of my laziness - I'm not very good at reading anyone else's stuff, if it extends to more than a page without equations in it.  I've always preferred to work things out for myself - an approach that has proved very successful for me in various aspects of life.  If I can't figure something out in my head, I scribble down a few diagrams or a balance sheet, and if that's not enough I build a little model.

These sort of models are intended to help me develop my own ideas about how such things work.  Ideally they should be stripped of everything other than the point I want to look at - as simple as possible whilst retaining the complexity to push my understanding a bit further.  The models I have included in some of my posts are examples of this.  They show things where I have a sense of how it works, but I need to see it in action to really get my head round it.

The type of model is driven by the issue I am concerned with.  But on the whole, I like models  that I can relate to the things we observe in a real economy.  I like to think that I could take any of these models and, maybe with a little tweaking, put some realistic numbers on them.  I wouldn't expect to get good forecasts or anything by doing that - it would all be about understanding how it really works.  Or doesn't.  Sometimes the conclusion is not what you expected.

I also look at more detailed models.  Again, I have on this blog some details on my UK macro model.  The purpose here is different, but not a lot so.  It is still about understanding.  A lot of it is simply the learning that comes through construction of a model.  Often simply having to organise and make sense of the data reveals important insights.  Otherwise the benefit comes from running experiments with the model.  With simple theoretical models, we are trying to understand a very specific mechanic.  With bigger, more complete models, the purpose is more general.  We are testing our intuitions, to see if everything works to together as we believe.  Often, this will draw out interesting effects that would have been hard to spot otherwise.  It's all about having a tool to aid our thinking - a grand version of a supply and demand diagram.

However, I think it is very important to recognise the limits to what models can do.  It is easy to get seduced into thinking that a model is some kind or oracle.  This is a mistake.  Any model is necessarily a huge simplification.  The results depend critically on the assumptions made.  However complex and detailed they are, all they really reflect is the theories of the modeller.

This doesn't invalidate the benefits I have talked about, but it means we must be careful how we use them.  They can help inform and quantify our judgements, but that is all.  If we don't understand the results, they are useless.  The model is not revealing any new truth, it is simply reflecting our own ideas, helping us to visualise how a massively complex system fits together.

I appreciate that there are economists have no interest in models.  That's fine - many of them I have great respect for, and the profession benefits greatly from having people take different approaches.  For me, the use of models is invaluable.  One of the things that most impressed me about Wynne Godley was his ability to combine economic theory with a deep insight into the real data.  His work in empirical modelling was key to that and it is the reason I do economics the way I do.

Wednesday, 11 September 2013

Banks, Non-Banks and the Interest Rate Effect

In my last post, I used the balance sheet framework of a simple imaginary economy to look at the relationship between bank lending, non-bank lending and payments money.  I thought it would be useful to use this framework to construct a little model.  I particularly wanted to look at the way non-bank lending and bank lending differ, even though both are expansionary.

A full list of the equations, variables and parameters for what follows is set out at the end of this post.

Many of the equations simply set up the accounting relationships shown in the balance sheet matrix.  This is the same as before and set out below:


Time deposits


Savings accounts



The financial sector is represented by two types of lender: banks and non-bank financial institutions (NBFIs).  Both entities make loans to firms (households are assumed not to borrow).  Banks offer time deposits to households and NBFIs offer something similar which we will call savings accounts.  The key difference between banks and NBFIs is that only banks can offer checking accounts.  As we are going to assume that only checking accounts can be used for payments, we're going to define money as the balance of such accounts.  Because they cannot offer checking accounts, NBFIs have to attract funds into their savings accounts before they can make loans.  As the timing of new loans does not necessarily coincide with the timing of new savings, they also hold some money balances as a float.

I am assuming that households behave as follows:
- they hold money in a fixed ratio to income;
- they have a long term target ratio of total financial assets to income, but move towards this incrementally; and
- they allocate their non-money assets between deposits and savings accounts on the basis of the rates paid on each.

Investment is driven by loans.  I am assuming the demand for loans exceeds supply, so the quantity of loans is determined by credit constraints imposed by banks and NBFIs.  The amount of lending by each can therefore be treated as exogenous.  As we are assuming firms do not retain any money, their spending must be equal to the change in loans.  

I am assuming here that neither consumer spending, nor investment depends on interest rates.  This makes it easier to isolate the effects we want to observe.  We could easily relax this assumption if we wanted.

This set up now means we can compare the effect of an increase in bank lending to an increase in NBFI lending.  The charts below show this for the same absolute increase in the value of loans made by each.

The first thing to note is that the effect on GDP is the same for bank lending as for NBFI lending (this depends on my assumption that spending is not affected by the change in interest rates described below).  New loans have a temporary effect on investment when they are made and a permanent, but lesser impact on consumption due to the increase in overall financial assets.

The difference between the two is in what happens to interest rates.  As we can see from the charts below, an increase in NBFI lending tends to push up rates offered to savers, compared with an increase in bank lending.

If they want to increase lending, NBFIs need to attract more savings.  To do this they will have to raise their savings rates.  This will tend to pull funds away from banks.  The total funds placed with banks cannot fall because it must always be equal to the total of bank loans, but what will happen is that banks will find the ratio of deposits to money falling.  Banks will react to this by raising their own deposit rates to try to attract more longer term funds.

Bank lending will tend to have the reverse effect.  Initially, bank lending adds to checking account balances and banks will need to raise their rates to try to attract savers into longer term deposits.  NBFIs will need to follow suit if they are not to lose funds themselves.  However,  once the initial boost from the new loans has worn off and income falls, household demand for money holdings also falls and they try to invest more into term deposits and savings accounts.  However, banks and NBFIs do not at this stage need more funds and so they can cut their savings rates.

To illustrate further what is going on, the graphs below show the deviations in the balances of household money holdings, deposits and savings accounts.

So we can see that even though bank lending and non-bank lending might both be expansionary, they are not equivalent.  The difference reflects their relative position in the hierarchy of money.  NBFIs need bank money to make payments.  Banks do not need NBFI claims.  In addition to facilitating debt funded expenditure, bank lending will also increase liquidity in a way that NBFI lending will not.  Banks and NBFIs are therefore different.  But this does not mean that NBFI lending does not matter.  


GDP is consumer spending plus investment by firms.

Y = C + I

Consumer spending and investment are defined by accounting identities.  Consumer spending is income less increase in assets (i.e. saving).

C = Y - ( V - Vt-1 )

Investment is equal to the increase in loans.

I = L - Lt-1

Households holdings of money are in a fixed ratio to GDP.

Mh = αm . Y

Households also wish to hold total financial assets in an fixed ratio to GDP, but this is achieved by a gradual adjustment based on the difference between a target and actual value.

V = Vt-1 + εv . ( V* - Vt-1 )

V* = αv . Y

The proportion of non-money assets that households wish to hold as savings accounts is a function of the interest rate differential between deposits and savings accounts.

S = ( V - M) . ( λh0 + λh1 . ( rs - rd ) )

Deposits are then the balancing item.

D = V - Mh - S

The amount of loans by banks and NBFIs adjust incrementally towards a target level.

Lf = Lft-1 + εf . ( Lf* - Lft-1 )

Lb = Lbt-1 + εb . ( Lb* - Lbt-1 )

Banks and NBFIs each have to make a portfolio decision based on rates.  The proportion of loans that banks would prefer to finance with money, as opposed to deposits depends on the deposit rate. 

M = Lb . ( λb0 + λb1 . rd )

The amount of money that NBFIs wish to hold depends inversely on the savings rate.

Mf = Lf . ( λf0 - λf1 . rs )

The above two equations are re-arranged as equations in the interest rate.  In addition to enabling solution, this reflects the idea that banks and NBFIs set the rates they offer and then take all money at those rates.

Lastly, some accounting identities.

M = Mf + Mh

L = Lf + Lb

Mf = S -Lf


Start Value
Consumer spending
Time deposits
Investment by firms
Loans by banks
Target loans by banks
Loans by NBFIs
Target loans by NBFIs
Total balance in checking accounts
NBFIs' balance in checking accounts
Households' balance in checking accounts
Savings accounts with NBFIs
Total household financial assets
Target household financial assets
Rate on deposit accounts
Rate on savings accounts


Household money holding ratio
Household wealth ratio
Bank loan adjustment rate
NBFI loan adjustment rate
Household wealth adjustment rate
Bank portfolio parameter
Bank portfolio parameter
NBFI portfolio parameter
NBFI portfolio parameter
Household portfolio parameter
Household portfolio parameter

The graphs show the result of an increase of 10 in the value of either Lf* or Lb*.