This is the second of three posts highlighting recent SSRN papers that point to the monetary base as the ultimate source of the business cycle.
In Part I, I discuss the defining feature of a monetary economy, namely that bank-financed demand creates its own nominal supply. Also, I trace the source of economic fluctuations to a disequilibrium between desired borrowings and savings. Here, I discuss the micro-foundations of desired borrowings and savings. Also, I put forth a new interest rate model to find the equilibrium conditions under which desired borrowings equal desired savings. In Part III, I discuss the implications for the business cycle.
First, I need to point-out the importance of micro-foundations. We cannot hope to understand the macro economy unless we understand how people make decisions. In my view, a macro system can be studied independently of its micro parts only if micro-interactions are random and largely cancel out. However, coordination and feedback loops on a micro level give rise to macro outcomes that are inscrutable unless you have realistic micro-foundations.
The keyword here is “realistic” micro-foundations. My SSRN paper, Redefining Utility in Terms of Gain-Loss Prospects, attempts to do exactly that. The paper reframes utility as an expression of human time and puts forth a new decision framework that is consistent with how people actually make decisions (I refer to the new framework as gain-loss utility; for highlights, please refer to this post: Bridging the Gap between Behavioral Econ and Mainstream Macro).
The critical decisions that affect the macro economy are decisions involving borrowings and savings. Specifically, we need to understand why people borrow and save and what factors determine asset allocation along the risk spectrum encompassing money, risk-free bonds and risky financial assets.
Gain-loss utility brings two key benefits. First, time preference and risk tolerance, the two aspects of human behavior that govern borrowing and saving decisions, arise endogenously within the utility framework itself. In contrast, mainstream macro uses exogenous parameters resulting in a number of puzzles such as the Equity Premium Puzzle and the empirical failure of the Euler Equation. Second, gain-loss utility is cardinal in nature, allowing for easy aggregation of individual decisions into macro outcomes. Accordingly, gain-loss utility can be used to derive the equilibrium level of interest rates at which aggregate desired borrowings equal aggregate desired savings.
According to gain-loss utility, an agent will borrow if she expects her income to grow faster than the geometric mean of her expected growth in needs and the risk-free interest rate. By the same logic, an agent will save if she expects her income to grow slower than the geometric mean of her expected growth in needs and the risk-free rate. Using this basic insight, a macro model can be constructed that reveals the necessary and sufficient condition for equilibrium between aggregate desired borrowings and savings. Aggregate desired borrowings and savings are in equilibrium when interest rates equal the aggregated income-growth expectations in the economy.
With respect to asset allocation, savers with negative income-growth expectations hold their savings in the form of money, giving rise to what I refer to as asset money demand. The rationale for asset money demand is simple. The interest rate on bank deposits cannot fall below zero due to the existence of paper cash. Zero is still higher than the negative income-growth expectations of such agents — by holding money, they engage in a risk-free arbitrage. They keep all the upside if future interest rates were to rise but none of the downside if rates were to fall because rates can never fall below their negative required returns [1]. Savers with positive income-growth expectations allocate their savings into financial assets in order to lock expected returns in excess of their income-growth expectations. In other words, they allocate away from money because the future interest rate on money is uncertain and may fall below their income-growth expectations, thereby preventing them from fulfilling their savings objectives.
As to risky financial assets, gain-loss utility shows that risk tolerance is given by the agent’s risky-asset ratio, which measures the investment in a risky asset as percent of the future income expected by the agent that’s unrelated to the investment at risk. The greater the risky-asset ratio, the lower the risk tolerance and the higher the return premium demanded by the agent for a given level of risk. Accordingly, borrowers issue risky assets until they are indifferent between borrowing at the risk-free rate or issuing risky assets instead. By the same logic, savers with positive income-growth expectations allocate into risky assets until they are indifferent between holding risky assets or risk-free bonds. In other words, borrowers and savers view risky assets on a risk-adjusted basis whereby the risk-adjusted expected return equals the risk-free rate [2]. Therefore, we can ignore risk and reduce financial assets down to two categories: money and risk-free bonds.
The only difference between money, in the form of interest-bearing demand deposits, and risk-free bonds is duration. Duration measures the change in the price of a financial asset or liability with a change in interest rates. Money does not have duration, meaning that its price does not change with a change in interest rates. Assets and liabilities with a fixed rate of interest and a term that extends beyond the current period have duration. Duration allows borrowers and savers to lock expected returns. Borrowers want to lock the interest rate on their liabilities to ensure it stays below their income-growth expectations. Savers want to lock the interest rate on their assets to ensure it stays above their income-growth expectations.
It is precisely the supply of duration by borrowers and the demand for duration by savers that give rise to capital markets. In other words, I think of capital markets not as the markets for loanable funds but as the markets for duration. Accordingly, interest rates are set in capital markets based on the supply and demand for duration.
Banks cannot play the role of market-maker in capital markets because bank liabilities (aka, money) have zero duration. As I discuss in Part I, the primary role of banks is to make a market for zero-duration borrowings and savings, thereby enabling the economy to bootstrap. When it comes to borrowings and savings with duration, capital markets rule. Every time a bank extends a loan at a fixed rate of interest, it creates a duration mismatch between its assets and liabilities that exposes the bank to interest-rate risk. Banks are not in the business of betting on interest rates but in the business of locking spreads. Furthermore, bank regulators require banks to duration-match their assets and liabilities in order to avoid interest-rate risk. For these reasons, banks sell duration in capital markets by either selling fixed-payer swaps or fixed-rate loans packaged into bonds. In other words, when it comes to assets with duration, banks do intermediate between duration sellers (borrowers) and duration buyers (savers).
And herein lies the problem. Remember asset money demand? That’s the long-term demand for money by savers with negative income-growth expectations. Because such savers do not participate in capital markets, the demand for duration (or alternatively, the supply of savings) in capital markets is insufficient for interest rates to converge with income-growth expectations. As a result, interest rates end-up higher than the equilibrium level, resulting in a shortage of desired borrowings. Only if banks were to go long on duration, can they create the money necessary to meet asset money demand and, in the process, lower interest rates to the equilibrium level. But we already know banks can’t go long on duration (or at least, they are not supposed to). This is where the central bank comes in (or gold under the gold standard).
The central bank is the only banking institution that can issue zero-duration money (aka, the monetary base) in exchange for long-duration assets. The central bank’s balance sheet is the only balance sheet that can withstand a duration mismatch. This means that the monetary base is the component of the money supply that can satisfy asset money demand [3]. Assuming that commercial banks perfectly duration-match, it is the monetary base that determines the level of interest rates relative to the equilibrium level. I have to stress that this interest-rate framework is quite different from either Loanable Funds or Keynes’ Liquidity Preference as illustrated on Chart 2 (for the formal model, please refer to the SSRN paper: Gain-Loss Utility, Equilibrium and Interest Rates).
I refer to the monetary base as exogenous because it is determined by a group of central bankers pursuant to some policy rule or discretion. In contrast, money created by banks is endogenous because banks simply respond to the demand for zero-duration borrowings. On a very high level, the impact of the monetary base on the business cycle can be described as follows:
- If the monetary base supplied by the central bank exceeds asset money demand, interest rates in capital markets fall below income-growth expectations in the economy, resulting in excess desired borrowings. In turn, this translates into excess aggregate demand that boosts nominal incomes, prices and employment (effectively, bootstrapping the economy toward full employment).
- If the monetary base is less than asset money demand, interest rates in capital markets rise above income-growth expectations, resulting in a shortage of desired borrowings. This causes a shortage of aggregate demand that depresses nominal incomes, prices and employment.
- If the monetary base equals asset money demand, the economy is in a steady state that may or may not be consistent with full employment.
It is not a coincidence that the most important exogenous input to the economy, namely the monetary base, should have such outsized influence on the business cycle. In Part III, I discuss the implications of this framework under three monetary regimes:
- Fixed reserves (gold standard),
- Variable reserves denominated in domestic currency (fiat money in advanced economies) and
- Variable reserves denominated in foreign currency (fiat money in developing economies).
[1] There is widely-held belief that returns on financial assets dominate the returns on money. Not true. For simplicity, let’s set risky assets aside. The yield on the 10-year Treasury is identical to the expected return on money over a 10-year time horizon, as projected by the 1-month forward curve. However, there is no guarantee that the 1-month forward curve will correctly predict the future path of interest rates on money. Only by buying the bond can a saver lock that expected return.
[2] Gain-loss utility lends support to the Efficient Market Hypothesis, according to which all excess returns are compensation for risk. The reason we talk about the Equity Premium Puzzle is because mainstream macro postulates an objective return premium for risk. According to gain-loss utility, such objective measure does not exist. Instead, the required return premium for a given level of risk is subjective to each investor based on his or her risk tolerance.
[3] Technically, central bank reserves, by far the largest component of the monetary base, are not included in the money supply because they are assets on commercial banks’ balance sheets. However, the corresponding commercial bank liabilities, which are counted in the money supply, can be viewed as monetized by the central bank.
