The current macroeconomic environment may appear complicated but is much simplified by focusing on two issues: First, the recent speculative boom and bust and the consequences following there from; Second the increases in spending (fall in "savings rate") over recent years and the current readjustment to more normal historical levels.
The current recession is in large part an adjustment of consumption and spending, as households raise their "savings rate" back to more historic norms. I believe the recovery may occur sooner, and be more robust, than many commentators are currently expecting.
The current crisis seems exceptional because speculative booms and busts have not been common since the second world war, but they were common in the US and Europe prior to 1929.
The current recession is in large part an adjustment of consumption and spending, as households raise their "savings rate" back to more historic norms. (See http://www.closemountain.com/papers/macro_persinc.html for most recent update) From the early 1990s to early 2008 the "savings rate" (simply income less spending divided by income) fell, from roughly 8% to virtually zero. This means that in the US macro economy as a whole, the total of aggregate income was spent. In addition, during the 2000s households took on increased levels of debt, primarily mortgage debt. Over the past few months the spending pattern has been sharply reversed, with consumption falling relative to income, the savings rate rising to 5.0% in January, and household debt falling.
The most likely explanation for the recent rebound in the savings rate and the changes in prior years is a "wealth effect" - a response of households to changes in wealth. The idea is that households consider total lifetime wealth or what Milton Friedman termed "permanent income". Changes in consumption or spending depend on changes in permanent income, not transitory income. This implies that an increase in permanent income, even when current or observed income does not rise, will produce an increase in current consumption. Similarly, a fall in permanent income will produce a fall in current consumption (and a rise in the savings rate).
My hypothesis is that sometime in the 1990s the US economy and the world economy experienced an increase in potential growth and thereby an increase in expected future growth and lifetime wealth or permanent income. (In addition, the volatility of GDP growth fell and overall employment increased. Both of these would have contributed to higher expected future earnings.) Households rationally increased current spending and debt during the 1990s and 2000s - i.e. decreased the savings rate. I just think households over-estimated potential growth and consequently lowered the savings rate too much.
With the current speculative bust expectations are being scaled back, estimates of future earnings (and thus wealth) are being reduced, and consumption is responding to the fall in permanent income. Starting in 2006/2007 real estate values started falling, in 2008 all asset values collapsed, and recently unemployment has soared. In combination this implies a substantial decrease in perceived lifetime wealth. In response we would expect households to decrease spending - i.e. to increase the savings rate.Economic Recession and Recovery
From April 2008 to January 2009 the savings rate (disposable income less spending divided by income) grew from 0% to 5%, and spending fell from 100% to 95% of income, levels not seen since 1998. This is a huge adjustment over a short period. Since consumption accounts for about 70% of GDP, a downward adjustment in consumption over a short period translates into a significant fall in GDP - a recession. Once the savings rate adjustment is complete, however, the growth rate of consumption can be the same as before the adjustment and the growth in GDP would be only slightly lower than before. (Furthermore, starting in the 1990s overall productivity in the US started growing at a faster rate than had been the case for the prior decades. This means that the potential for growth going forward may be slightly higher than it was during the 1970s and 1980s, possibly offsetting any small effect of a lower spending-to-income ratio.)
The large increase in savings rates over the past few months means that households have already accomplished substantial adjustment. It is difficult to predict how much adjustment will be necessary, but I would guess that the savings rate may stabilize at around 8%. If this is correct then more than half the adjustment has been accomplished. The bottom line is that the current downward adjustment in household spending goes far to explaining the recession, but once the adjustment is complete the economy can grow again. There is a chance that the recovery will occur sooner, and be more robust, than many commentators are currently expecting.Current Recession and Stimulus Package
My theory is that the recession is primarily due to households adjusting to new levels of permanent income (in response to earlier over-estimates and the bursting of the speculative bubble), and that we are experiencing a period of painful but necessary adjustment. Further, when households have adjusted then consumption growth, and GDP growth, will resume. The recent US stimulus package, in contrast, is predicated upon the assumption of "deficient demand" based primarily on experience from the Great Depression. It presumes that the economy is fundamentally broken, with consumer spending permanently impaired, and that only government borrowing and government spending can remedy the situation.
If my idea is correct then the economy will correct of its own accord so that the stimulus package is not really necessary. (What is necessary, and seems to have been accomplished, is to ensure that the financial system does not collapse and bring down the rest of the economy.) Going further, however, the stimulus package could do active harm, by increasing government debt, future tax burdens, and thus further decreasing households' perceived wealth.
We are in a credit or liquidity crisis that seems to be following the pattern of 19th century speculative booms and busts described for example in Charles Kindleberger, Manias, Panics, and Crashes. (See below and http://www.closemountain.com/papers/macro_liquid_0901.pdf for more detail.) The current crisis seems exceptional only because speculative booms and busts have not been a prominent feature of the economic landscape for the the US and Europe since the second world war. It would no doubt be more familiar to someone living during the late 1800s and early 1900s when speculative booms and busts recurred periodically. For the US Kindleberger cites five speculative cycles over the period 1860-1940: peaking in 1873 (speculation in railways, homesteading, and Chicago building), 1892-93 (silver and gold), 1907 (coffee, Union Pacific railway), 1920-21 (securities, ships, commodities, inventories), and 1929 (land to 1925, stocks 1928-29), and an additional four which did not involve the US but variously affected Britain, Italy, and France.
The reaction of investors, consumers, and companies to the debt deflation phase explains much of the current macroeconomic environment: Falling Asset Prices; the Impact on Banks and the Financial System; the current wave of ForeclosuresFalling Asset Prices and Complex Securities
Much has been written about the complex securities (CDOs, CDS, ABS, etc.) and how they have contributed to the current crisis. While possibly true at the margin, the real story, to my mind, is nothing more complicated than a speculative boom followed by falling asset prices. As Kindleberger lays out, during the debt deflation phase of a speculative boom / bust cycle, asset prices fall across the board. In the current crisis the price of houses, stocks, corporate bonds, oil, iron ore – in short the price of virtually any risky asset has fallen. This is not a complicated story, but rather one that has been repeated over the millennia: a speculative boom starts in a single asset, spreads to other assets, and is followed by a precipitous fall in the prices of all risky assets. Tulip bulbs, South Sea Company stock, canals, commodities, exports to Brazil, cotton, railroads, wheat, securities, land, have all been the focus of speculation at one time or another. And during the speculative bust, asset prices have fallen across the board.
The current crisis is not principally the fault of complex securities (although they no doubt contributed) or a nefarious financial system (again that no doubt contributed) but rather the result of simple human credulity and miscalculation. House prices have fallen because house prices rose to unsustainable levels. Banks and investors lost money because they made unwise investments based on house prices continuing to rise, and they have now fallen. There is nothing complicated in losing money when asset prices fall.Impact on Banks and the Financial System
Falling asset prices have a very simple impact on banks and the financial system. With the onset of the debt deflation phase of a speculative bust everyone craves liquidity and moves away from risky assets. This means a dramatic increase in liquidity preference or demand for money – households, companies, banks, everyone wants to hold liquid assets. In past crises in the US that I have looked at (late 1800s and early 1900s, see http://www.closemountain.com/papers/macro_liquid_0901.pdf) the sharp increase in demand for money has led to falls in the general price level and major banking or monetary disturbances, due at least in part to the failure of money supply expanding in concert with money demand. The Great Depression following 1929 was a particularly virulent episode of falling prices and banking collapse. As asset prices and the general price level fell, bank loans went bad, banks failed, commerce slowed, and the Great Depression was born.
The Federal Reserve and Treasury have, with the TARP and related programs, provided reserves to the banking system, helped re-capitalize banks, provided guarantees to the money-market mutual fund industry, and generally expanded the money supply on a massive, unprecedented scale. These seems to have (for the moment at least) forestalled collapse of the US banking system and prevented runs on the financial system.
The TARP program has been criticized as ineffective because lending has not returned to the pre-2008 levels. This criticism is misplaced for two reasons: First, too much lending was a major contributor to the speculative boom and it hardly seems a good idea to re-expand the speculative bubble. Second, the TARP program was not intended to expand lending to pre-crisis levels but rather to avert complete collapse of the banking sector, which it has done effectively. In this sense the TARP program is the classic lion that didn’t roar – complete success is measured by the failure of anything to happen.Foreclosures
I believe the archetypical foreclosure today is quite different from that of the past. When the house price is below the mortgage (“negative equity” on the mortgage) the economic incentives for the homeowner are quite different than when the house is worth more than the mortgage. The current housing crisis is the first time in decades that house prices have fallen broadly and deeply and I believe that many “foreclosures” today are closer to defaults by the homeowner than true foreclosures by the bank. Understanding the how and why of this illuminates the current surge in foreclosures and the likely efficacy of efforts to stem the foreclosure problem.
Historically homeowners have been very reluctant to default on a mortgage and generally made strong efforts to pay the mortgage even in severe circumstances. There is a strong economic incentive against default when the home is worth more than the mortgage: defaulting on the mortgage means losing equity in the home. There is a secondary incentive for continuing to make monthly mortgage payments rather than rental payments, in that part of the monthly mortgage payment goes towards building home equity and the remainder is deductible against income taxes. (As an example consider a $100 mortgage on a $120 house. If a homeowner is foreclosed against they will lose some or all the $20 equity. The monthly payment would be roughly $0.57 for a $100 mortgage (the rate for a 15-year fixed-rate mortgage is roughly 4.76%) with $0.17 going to principal and thus increasing the homeowner’s equity and $0.40 deductible against income.)
In contrast there is a strong economic incentive for the homeowner to walk away from the mortgage (default) when the house is worth less than the mortgage: default increases the homeowner’s net worth. When the house is worth less than the mortgage (“negative equity”) the homeowner’s net worth is negative in that the homeowner would take a loss on any sale of the property. Walking away from the property and letting the bank take possession transfers the loss from the homeowner to the bank. There is still the tax deduction on the interest portion of the monthly payment, but now the payment of principal benefits the bank rather than the homeowner. (Again consider the example of a $100 mortgage but say the property is worth $80. If the homeowner is foreclosed against they will gain $20 because they lose an $80 asset but renounce $100 of debt. The monthly payment is still $0.57 but now the $0.17 principal payment goes straight to the bank with no increase in the homeowner’s equity.)
There are clearly costs to a homeowner when defaulting on their mortgage, most obviously the negative credit history. But eventually, as the value of the house goes down, the costs may be outweighed by the benefits of defaulting on the mortgage. Just as important, a negative equity position means that a homeowner has no cushion of wealth built into their home upon which to rely during difficult circumstances. (Consider any temporary loss of income or increase in expenses, say a temporary job loss or a medical emergency requiring cash. When the house is worth more than the mortgage the homeowner can use the positive equity as a cushion and re-mortgage the house to meet the temporary crisis; with negative equity there is no cushion and default may be the only option.)
If I am correct in arguing that the current wave of foreclosures is driven by homeowner default as much as by bank foreclosure, then many of the current efforts to lower foreclosures will be ineffective. Reducing monthly payments will not have a substantial impact upon economic incentives for a homeowner with negative equity, because the main cost of continuing with the mortgage is the negative equity or negative net worth, not the monthly payment.
If the current foreclosure crisis is being driven (in part) by homeowners rather than banks, this goes far in understanding why banks are in such poor shape today. Under this hypothesis homeowners have transferred losses from falling home values to banks rather than retained those losses. This is certainly consistent with the broad picture of bank troubles and consumer spending. Banks started feeling distress in mid-to-late 2007 when house prices first started to stagnate and fall. Consumer spending did not contract substantially until much later, the fourth quarter of 2008. This is what one would expect if consumers were able to transfer some of the losses due to falling house prices to bank shareholders and bondholders.
home page | Contact Close Mountain Advisors