The Growth of Offshore Corporate Cash Pools

The European commission is attempting to claim back tax on billions of profits made by Apple through its Irish operations.  This story is interesting in itself in illustrating the issue of tax base erosion and profit shifting in an increasingly globalised world.  However, it also has macroeconomic implications that go beyond the effect on government tax revenues.

Apple is just one of many US parented[1] multinational corporations that have organised their affairs so as to be able to pay a very low effective rate of tax on profits earned outside of the US.  In many cases this has been accompanied by a shift of profits out of the US, either as a result of relocating productive activity or, more controversially, by things like the choice of location of intellectual property rights.

The profits from these operations, which have been substantial, must then be retained offshore to avoid a significant US tax charge which would arise on repatriation.  This has led to the accumulation of large pools of liquid assets held by corporate entities outside of the US, generally denominated in or hedged into dollars.  Apple's last 10-K showed $147bn of cash and available-for-sale securities, much of which is likely to be in non-US entities.

Looking at sectoral flows and balances, one notable feature of recent times has been the reversal of the standard model in which households save and firms borrow.  The recent trend has been for falling household surpluses whilst non-financial corporates have become net savers.  Globalisation and the incentive that creates to accumulate offshore cash piles has been part of the reason for this.

If these funds were to be repatriated they would appear as current account income for the US.  Instead, by accumulating dollar assets, they help fund the deficit.  So the process by which US production and profit has been moved offshore, contributing to the size of the deficit, has at the same time contributed to funding it.

This rise of large corporate cash pools has had further implications.  With the substantial amounts involved, these investors are increasingly looking to diversify their credit risk.  Placing all the money on deposit, even spread around between different banks, would result in some massive exposures.  So investors like this are hungry for deposit substitutes, particularly those that are collateralised.  Securitisation allows the financial sector to create the assets suitable for use as collateral, so this demand for secured on-demand claims helped feed the growth of shadow banking[2].

Understanding the way that multinational corporates have structured themselves helps shed some light on certain international and intersectoral flows, some of which have had important macroeconomic consequences.  The international tax system has been a key element in shaping these structures.

[EDIT - As Ramanan has pointed out to me, undistributed earnings of all foreign associated enterprises are treated as distributed and re-invested.  Retention, rather than distribution, therefore has no impact on the current account, but rather results in two offsetting movements in the financial account.  Movement of productive activity would therefore affect the balance of trade, but not necessarily the current account.]

---

[1] This applies to other jurisdictions as well, but it is most notable in the case of the US.

[2] See Pozsar, Z. (2011) Institutional Cash Pools and the Triffin Dilemma of the U.S. Banking System for more on this.

Stock Flow Ratios and the "Velocity of Collateral"

I've read a couple of good blog posts in the last few days which, although apparently unrelated, have an interesting connection.

The first is Brian Romanchuk's piece in which he gives a nice, clear account of the role of stock-flow norms in economic modelling.  He emphasises the importance of distinguishing between stock variables and flow variables.  Stocks represent the state of affairs at a specific point in time; flows are what occur during a period of time, or simply between two specific points of time.  Just as it is important to know what variables are stocks and what are flows, it is also important to distinguish different types of ratio: flows to flows, stocks to stocks, or stocks to flows.

The second is Scott Skrym's post on the "velocity of collateral" - a term used to highlight the way collateral is re-used in repo and other transactions, so that the same securities can be posted several times in so-called "collateral chains".  Skrym provides a good description of the way this works and, as always, some useful context from current trends in the repo market.

People who use the term "velocity of collateral" like to present it as analogous to the velocity of circulation of money.  We have a stock of eligible collateral, rather than a stock of money, and in each case we have a certain volume of transactions that the collateral, or money, is used in.

However, there is a confusion between stocks and flows that is creeping in here.  When the velocity of circulation of money is considered it is in terms of comparing the stock of money with a measure of flow.  Typically that flow is the monetary value of the transactions in a given period.  So we take a period of, a year say, and add up the value of each transaction that has taken place within that year. 

The unit of measurement for this value (being a flow) will be dollars per year.  In comparison, the unit of measurement for the money supply will be just dollars.  So when we divide the value of transactions by the stock of money, we get a measure of velocity of circulation which is expressed as number of times per year.  We can interpret this as being the number of times each dollar changes hands per year, on average[1].

When calculating the "velocity of collateral", however, the stock of collateral is compared with the volume of collateralised transactions outstanding at any given time.  This latter variable is a stock concept.  It is measured as a pure dollar value, not a value per unit of time.  This means that when we divide by the quantity of collateral, we get a number expressed as a pure ratio, not as a number per unit of time.

"Velocity of collateral" is a stock / stock ratio; velocity of money circulation is a stock / flow ratio.  They are very different concepts.  That is not to say that the thing that "velocity of collateral" measures does not matter; rather that we need to be wary of interpreting it as being comparable to the velocity of circulation of money.

---

[1] Although, as I have written previously, we should be careful with this interpretation.

Own Rates of Interest and Real Rates of Interest

David Glasner had a couple of posts recently (here and here) considering real rates of interest in a barter economy.  I'm not sure I quite agree with him even in his last post on the topic, so I thought it was interesting to look at further.  I'm going to use the numbers from his example.

The issue he is looking at is the pricing of loans in a barter economy.  In this economy, loans have to be constructed with commodities.  So I might lend you 100 onions at 5% interest, which would mean that at the end of the loan you would pay me back 105 onions.  We are assuming this is a proper loan, like a loan of money, so onions are simply the denomination and the settlement medium.  In other words, you do not have to return to me the same onions I loaned you, plus five more.  You simply deliver to me 105 onions of the required type.

We can say that 5% is then the "own rate" for loans of onions.  We then consider what own rates might apply to loans of other commodities, like tomatoes or cucumbers.  For this, we assume that normally onions, tomatoes and cucumbers all trade at par with one another, but currently there is lower demand for tomatoes and higher demand for cucumbers.  So, at the current prices, 100 tomatoes exchange for 90 onions and 100 cucumbers exchange for 110 onions.

Simple arbitrage then dictates what the own rates of interest must be for tomatoes and cucumbers.  For example, someone might borrow 900 onions and exchange them for 1,000 tomatoes.  To repay their onion loan they need 945 onions (principal plus 5% interest), which they can get at the end of the loan by exchanging 945 tomatoes at the then par rate.  So a loan of 1,000 tomatoes will require repayment of 945 tomatoes, an interest rate of -5.5%.  Anything else will allow endless profits from borrowing in one commodity and lending in the other.  A similar argument shows that the own rate on loans in cucumbers must be 10.5%.

We need to ask how there can ever be a negative own rate of interest, such as we have here on tomatoes.  Does this not mean that someone can borrow 1,000 tomatoes, repay 945 and walk away with 55 tomatoes in profit?  The answer relates to the time aspect of the loan and the implications for storage.  If tomatoes are perishable, it may not be possible to store them from one period to the next at all, so this arbitrage may not be available.  Even if we take a non-perishable commodity, such as a precious metal, there may be storage costs such as security and these may eliminate the potential profits.

It is worth considering what would happen if there were no storage costs.   People would then want to borrow more tomatoes in order to simply hold them and take the profit.  This would tend to bid up the own rate on loans of tomatoes, which would prompt other people to exchange cucumbers and onions for tomatoes to be able to make more tomato loans.  This in turn would increase the current price of tomatoes in terms of these other commodities.  So both the own rate on tomato loans and the current price of tomatoes increases.  This continues until the arbitrage is eliminated.

So we have a different own rate for each commodity, but the relationship between all the rates is tied to the price structure.  With prices given, if one rate changes, they must all change.

If we introduce money to this economy, it must fit the same structure.  There will be a money price for each commodity.  We can work out the appropriate rate for money loans by reference to the current and future price of onions and the own rate on onions in order to meet the no arbitrage condition.  If we do the same exercise using tomatoes or cucumbers, we will get the same rate.

The question now is what the real rate of interest is here.  We have a different own rate of interest on each commodity.  Which if any is the actual real rate of interest?  In fact, we cannot say what the real rate of interest is without specifying the commodity in which we are expressing it.  There is no absolute real rate of interest that can be expressed purely in terms of time value.

That is not to say that time value does not matter.  If the time preference for consumption of vegetables changes, all of the own rates will change.  It is simply that we cannot point to a single own rate and say that this is the one that reflects pure time value, abstracted away from relative price movements.

In practice, we determine real rates of interest by reference to baskets of commodities.  We often calculate a real rate of interest using a consumer price index.  If we were to use a different index we would get a different result. 

We can also use baskets of commodities in our imaginary economy.  For example, we could calculate the own rate on a basket of 10,000 onions, 20,000 tomatoes and 30,000 cucumbers.  In this case, the own rate works out at 6.75% (i.e. you get back 10,675 onions, 21,350 tomatoes and 32,025 cucumbers).  This means that we would be indifferent between lending the basket at 6.75%, or lending in any of the individual commodities alone at their respective own rates.

As with the individual commodities, the appropriate rate depends on the basket we choose - there is no true rate that is independent of that choice.  However, whatever basket we choose to reference, the arbitrage free rate required on money loans will always be the same.

Making Sense of Liquidity Preference

Philip Pilkington has a post on liquidity preference.

I've never liked the textbook version of liquidity preference.  This has the interest rate as being determined through the choice of investors between bonds and money.  It kind of makes sense if you imagine a world with only interest bearing bonds and non-interest bearing money, but it's harder to interpret when you want to factor in banks and bank money, which generally pay some form of interest.

In Pilkington's version, instead of money we take the collection of assets that serve as cash equivalents, e.g. short-term deposits, Treasury bills, money market fund shares, reverse repo, etc.  These will generally be interest bearing and although the individual rates will differ, we can take a single short rate as representative of the return on these assets.  On the other side we have longer term financial assets including government and corporate bonds.  Again, we can take some long term rate as representative of the interest rate on this class of asset.

We thus have two asset classes and two representative interest rates.  Changes in demand (and supply) between the two asset classes can then be invoked to explain, not the absolute level of rates, but the relative rates.  In particular, we can take the short rate as being set by central bank policy, and then take the long rate as being determined by relative demand and supply for short and long assets.  Since the collection of short term assets are often described as being a kind of liquidity pool, we can think of the relative demand for short versus long assets as being a matter of liquidity preference.

This formulation makes more sense to me.  However, whilst it may be useful to understand what drives markets, as economists we may be more concerned with how this might impact on actual demand in the real economy.  When looking at how financial shocks feed through the system, there may be more going on than can be easily described in this model.  Some points are worth making.

Long term interest rates on corporate debt might matter, but they're not the be-all-and-end all.  It may be that the short term rate has a greater impact on real activity.  And, although that may include expectations of future short rates, we need to avoid confusing what causes changes in the long rate.  A change in the long rate due to a change in preference for long-dated assets is not necessarily the same as a change due to a change in expectations about future short rates.

   

We have to be careful talking about liquidity as the term is not used consistently or, at least, there are different forms of liquidity.  So whilst it can be useful, in a certain context, to think of short term assets as representing liquidity and longer ones as not, that's not always the way it breaks down.  More commonly, the liquidity of an asset is defined as the ease with which it can be sold without affecting the price.  By this measure, long dated government bonds may be more liquid than short-dated corporate obligations (they are treated as such for the purposes of the BIS Liquidity Coverage Ratio).

The divergence between short term rates and long term corporate debt rates reflects a number of things, including the term structure of interest rates and credit spreads, as well as the actual liquidity of the asset.  These do not necessarily move in the same direction, nor do they necessarily have the same impact on demand.   

The causal chain from financial market movements to spending is not only about interest rates.  Financial frictions and institutional structure mean that quantity constraints may be more important.  Lenders may partially respond to changes in funding cost by adjusting their credit criteria, thereby mitigating the effect on lending rates.

In general, we have a whole range of financial assets, each carrying different elements of term structure, credit risk and liquidity.  In addition, we have a variety of financial institutions with different regulatory and commercial structures, which also face certain financial frictions, preventing instant adjustment to new equilibriums.  Sometimes, dividing assets into two distinct classes to illustrate liquidity preference is a useful exercise and can help explain things we see going on.  Sometimes, it's more complicated and simply looking at two classes of asset is inadequate.  In my view, much of what happened in the crisis is difficult to explain purely within the terms of a liquidity preference model.  Either way, understanding the structure of interest rates as the result of demand and supply pressures for different types of asset provides a useful framework for this sort of issue.