Should the "natural rate" of interest be zero?

The argument for a zero risk-free rate goes as follows:

1. It is immoral to allow a rentier to earn a risk free return. A nonzero Fed Funds rate allows this.
2. The Loanable Funds model is incorrect: the interest rate does not balance savings and investment
3. Existing Taylor rule type techniques are not very good.
4. The government needs to be in a net deficit position in order to meet the savings needs of the private sector. If this is the case, then the call money rate will fall to zero unless the government "supports it" by draining cash.
5. Therefore we should not drain cash and let the interest rate fall to zero.

My initial take on this is #5 does not follow from 1-4. The rate should be zero only if it overall welfare maximizing, and the rate should not be zero if it is not overall welfare maximizing.

In this post, I'll look at assertion #1.

I think this is basically a confusion about how the financial markets work. It's helpful here to disaggregate between the domestic private non-financial and domestic private financial sectors (e.g. "banks" and "investors").

The following charts the real MZM Own Rate and the real FedFunds rate:

The MZM Own Rate is the asset-weighted return earned by deposits immediately available for withdrawal. These are primarily (but not always) government insured, and so this rate is a good proxy for gauging the return available to investors in exchange for not parting with liquidity and not incurring risk. That return has averaged -1.89% since the start of the time series. There is some evidence that when the FedFunds was too high and when the FedFunds was too low, that the real return was positive. As MZM will be required to pay something to depositors, as the deposits do have a nonzero value to banks, a zero FedFunds rate would guarantee a positive risk-free return, whereas a non-zero FedFunds rate is generally associated with a negative risk-free return.

Perhaps the FedFunds rate is somehow causing returns to be higher than they otherwise "should" be, if it were zero. I think this is a confusion about what drives returns.

The long term return on equity (dividends + capital gain) is the GDP growth rate. However, there is a lot of volatility and a long term capital commitment is required to be able to reliably obtain this return. In exchange for a reduced level of capital commitment and less volatility, investors relinquish some yield and purchase bonds or other instruments. The reduction in yield is a function of changing investor preferences. Investors do not have the option of accessing money at the FedFunds rate. As the level of capital commitment decreases, the rate of return declines to the level of MZM, not to the FedFunds rate. The latter is an overnight rate reserved for banks.

Banks, on the other hand, do have the option of earning (on their excess reserves) the overnight rate, and of course they must also pay that rate. Here, there are no "moral" issues, since we are not talking about rentiers extracting money from productive society, but of bankers extracting money from each other. The purpose of the rate is to put a floor under the banks' cost of capital.

But still, aren't those costs passed onto the borrowers, who are forced to pay higher interest rates? Isn't this a moral issue?

That depends on whether you believe that there is a moral imperative for borrowers to have cheap access to loans. I will get back to that. In any event, banks will always charge a premium to borrowers as some loans will not perform, and the reality is that neither banks nor regulators can predict the tail end of distributions. Therefore the FedFunds rate serves as an additional safety valve, forcing the banks to be somewhat more conservative in their loan loss estimates than would be the case if their cost of capital were lower. But the difference is slight. If the banks cost of capital is 5%, and the bank has a capital requirement of 12:1, then $1 of capital can still fund $12 of loans, and this means that the bank must charge a premium of at least .05/12 or 0.4% more than it otherwise would if the FedFunds rate was zero. But if only $2 were borrowed, then the bank would need to charge a premium of 2.5%.

But herein lies the rub. Bank lending is driven by the consumers demand to borrow, not the bank's desire to lend. So the number of loans per dollar of capital changes, and if the cost of capital were to remain constant, then this premium would shift, and this would be disruptive to consumers. So this rate should fall when the demand for loans falls, and it should increase when the demand for loans increases, in order to keep the premium per loan constant.

At the same time, the safety valve itself may need to change due to changing collateral values and charge off rates.

The FedFunds rate is a floor on the banks' cost of capital, which can be thought of as a function of the FedFunds rate + various risk premia. If those risk premia increase, then in order to keep the loan cost to customers constant, the FedFunds rate would need to fall. If those premia decrease, the rate would need to rise.

All of the above are arguments that the FedFunds should be variable if lending rates are to be managed. But should they be managed? Perhaps we should let lending rates rise and fall based on the market's willingness to fund bank capital and willingness of borrowers to borrow. Here there are two key insights

• changes or shocks to the interest rate can be much more harmful than the rates themselves, primarily the risk is in upward spikes of borrowing rates, meaning that the government needs a tool to quickly lower the rates paid by customers much more than a tool to keep the rate low.
• The Level of interest rate plays a key role in determining the relative prices of a narrow set of assets within the economy, particularly houses. When it comes to the bulk of assets for which loans are made, there is no economic benefit to having the interest rate be high or low, other than the rate should be held constant or adjusted to "smooth" demand for the item.

Bank lending is primarily a real estate phenomena -- the Schumpeterian hero that borrows in order to expand capacity and innovate does appear in history, but is a statistical discrepancy in the flow of funds. The majority of productive investment is funded by retained earnings. Almost all loans are issued for: residential real estate, condos, apartment complexes, shopping centers, etc.. To a lesser degree there are loans for the purchase of consumer durables and college tuition.

The key thing to notice here is that the vast majority of loans are made on assets that do not produce a stream of cash-flows at all, and there is deep ambiguity about how to "value" a house, car, or a college education. Therefore the credit available to fund these expenditures, given a level of demand, will determine their price. So if you believe that there is value in smoothing demand for these assets, and that there is a social cost to be born by these assets being expensive relative to other assets in the economy, then you should believe in a non-zero bank cost of capital that adjusts in opposition to borrowing demand.

The Fundamental Asymmetry

At the micro-level, there are incentives to increase productivity and decrease unit labor costs, paying workers less for each unit of output. But the costs of this -- decreased incomes to purchase inventory, are spread around. The only "incentive" that the market has to offer is an overall decline in sales in the form of a demand-led recession. There is no individual incentive to raise unit labor costs.

With the introduction of debt-financing, if accompanied by falling interest rates, even that asymmetrical incentive can be delayed, ushering in long booms of high profits, low inflation (since unit labor costs are low), and a significant deterioration in wages and an explosion in household borrowing. The main mechanism of household borrowing is for housing, and to a lesser degree, education -- both things that seem like assets but do not have a directly observable return. Once borrowing for these ends, the economy finds itself in a free-fall as businesses can no longer clear inventory.

If government stimulus spending is oriented towards purchasing inventory -- being in effect the consumer of last resort, then the key issue -- low wages -- will not be addressed and the result can be increasing unemployment even as total sales increase. If government spending is oriented towards boosting wages, either with a payroll tax holiday or directly employing the unemployed at high wages, then unit labor costs can be raised and workers can get back into a situation in which inventory clears without the need for households to take on more debt.

Lincoln Senate Address on Monetary Policy

From Kansas City

Money is the creature of law, and the creation of the original issue of money should be maintained as the exclusive monopoly of national government. Money possesses no value to the state other than that given to it by circulation.

Capital has its proper place and is entitled to every protection. The wages of men should be recognized in the structure of and in the social order as more important than the wages of money.

No duty is more imperative for the government than the duty it owes the people to furnish them with a sound and uniform currency, and of regulating the circulation of the medium of exchange so that labour will be protected from a vicious currency, and commerce will be facilitated by cheap and safe exchanges.

The available supply of gold and silver being wholly inadequate to permit the issuance of coins of intrinsic value or paper currency convertible into coin in the volume required to serve the needs of the People, some other basis for the issue of currency must be developed, and some means other than that of convertibility into coin must be developed to prevent undue fluctuation in the value of paper currency or any other substitute for money of intrinsic value that may come into use.

The monetary needs of increasing numbers of people advancing towards higher standards of living can and should be met by the government. Such needs can be met by the issue of national currency and credit through the operation of a national banking system. The circulation of a medium of exchange issued and backed by the government can be properly regulated and redundancy of issue avoided by withdrawing from circulation such amounts as may be necessary by taxation, re-deposit and otherwise. Government has the power to regulate the currency and credit of the nation.

Government should stand behind its currency and credit and the bank deposits of the nation. No individual should suffer a loss of money through depreciation or inflated currency or Bank bankruptcy.

Government, possessing the power to create and issue currency and credit as money and enjoying the right to withdraw both currency and credit from circulation by taxation and otherwise, need not and should not borrow capital at interest as a means of financing government work and public enterprise. The government should create, issue and circulate all the currency and credit needed to satisfy the spending power of the government and the buying power of consumers. The privilege of creating and issuing money is not only the supreme prerogative of government, but it is the government's greatest creative opportunity.

By the adoption of these principles, the long-felt want for a uniform medium will be satisfied. The taxpayers will be saved immense sums of interest, discounts, and exchanges. The financing of all public enterprises, the maintenance of stable government and ordered progress, and the conduct of the Treasury will become matters of practical administration. The people can and will be furnished with a currency as safe as their own government. Money will cease to be the master and become the servant of humanity. Democracy will rise superior to the money power.

Abraham Lincoln, Senate document 23, Page 91. 1865.

Good Trade and Bad Trade

If we are off-shoring low-value add activities, and the freed up labor is used for higher value-add activities, then compensation would rise. Conversely if we are off-shoring high-value add activities, and the labor is freed up to work as Walmart greeters, then compensation would fall.

More at The Other Cannon.

More Chart Porn: Household Borrowing as a recession predictor

Almost as good as the yield curve, and smells much nicer. From the Flow of Funds.

Unit Labor Costs and Inflation

For those worried about hyperinflation, it's good to keep in mind that inflation is not caused by "printing money" in and of itself, or by an increase in wages. You need high wages per output, or high unit labor costs (ULC):