Signs indicate that the stock market is on the way up, although bank failures now are up to 146 year to date, ahead of last year. Financial advisors will as always have to nudge investors towards taking a longer-term view, since the human tendency is to overvalue the immediate moment, resulting in what behavioral economists call “hyperbolic discounting.”
Hyperbolic Discounting By Stephen J. Huxley, Chief Investment Strategist, Professor of Business Analytics, Univ. of San Francisco
We hear a lot about the differences between institutions and people when it comes to investing. This is a good thing because there are significant differences between the two in terms of decision-making and investment behavior. In fact, a Nobel Prize in Economics was awarded in 2002 to one of the early pioneers of Behavioral Economics, Daniel Kahneman, a psychologist by training.
One of the terms used in conjunction with behavior economics has an intriguing name: “hyperbolic discounting.” What is its meaning and relevance to financial advisors who are on the front lines when it comes to dealing with people and their financial decisions?
The first thing to understand about hyperbolic discounting is that it is a bad thing. It attempts to put a mathematical name to behavior all advisors have observed: unless restrained, people will tend to put too high a value on “now” and too low a value on future events. This makes their investing preferences inconsistent over time.
Psychologists typically note this inconsistency in addiction behavior. A smoker may agree that the best plan is to enjoy smoking today, but to quit tomorrow in order to get health benefits. What he is saying is that for today, smoking has higher value than health benefits but, beginning tomorrow, health benefits will have a higher value. He is discounting the value of the health benefits so much that they have a lower value today than smoking has today, so he smokes today.
The next day, however, when tomorrow becomes today, the same thing happens. He will again enjoy smoking “today” and quit “tomorrow.” If this continues, the smoker will never give up, even though he claims he wants to quit. Hence, he suffers from time inconsistency – his tradeoff ratio between current and future actions does not stay constant. It changes with the passage of time. The point is that “now” has especially high value compared to any future time, whenever “now” is. The root problem is self-control; other examples would include dieting, procrastination – and saving for retirement.
It is normal, of course, to enjoy gratification sooner, rather than defer it. Indeed, that has always been true, because people naturally have a positive time preference – they want good things now and prefer to postpone bad things to later. But investing rationally means making this tradeoff consistent over time.
Mathematicians have formulated the discounting of future values in two ways: exponential and hyperbolic.
Exponential discounting leads to the classic present value formula that all institutions and markets use. Time is modeled as an exponent in the denominator: PV = FV/(1+r)^n, where r is the rate of interest and n is the number of years. As n gets bigger, PV gets smaller, i.e. is discounted exponentially. Furthermore, this discount ration would stay the same whether discounting next year to this year, or discounting the Year 21 value to Year 20. It is time consistent.
But with hyperbolic discounting, time is entered as a multiplier, such as PV = FV/(1+n). This does a better job mathematically to generate more rapid discounting of the near future (next year’s FV would be 50% of today’s) but less rapid discount of the distant future (Year 21 would be worth 95% of Year 20’s value). It is this difference that makes hyperbolic discounting behavior inconsistent over time. Any strategy that helps clients overcome this natural tendency – to use exponential discounting rather than hyperbolic discounting – is desirable.
Barclays Capital Economics Research Outlook: United States
The consensus of economists sees no significant improvement in growth in the near term. The November Bloomberg survey shows real GDP growth of 2.2% in both Q4 10 and Q1 11, a little better than the 2.0% reading in Q3 10. In contrast, we see growth of 3.0% in Q4 10, and 2.5% in Q1 11, and, if the tax cuts set to expire at year-end are extended, we would be likely to increase our Q1 11 forecast. Data on September trade and wholesale inventories led us to raise our tracking estimate of Q3 10 real GDP growth to 2.3%, up from the advance estimate of 2.0% and the Q2 10 reading of 1.7%. As discussed last week, the October employment report, ISM indexes, and auto sales all provided evidence of a further pick up in growth. This week brought a 24k drop in initial jobless claims to 435k. Given its volatility, we do not to put much weight on any one weekly reading, but the more stable 4-week average dropped to its lowest reading since September 2008. Figure 1 shows that initial claims (inverted) and real GDP growth are correlated, and the drop in claims so far in Q4 10 is consistent with the improvement in growth that we expect.
The stock market agrees a turn is coming.
While it is by no means perfect, the stock market is a good leading indicator of growth, typically leading growth by about a quarter. The reading on the stock market so far in Q4 10 is consistent with the idea that growth is set to turn higher. We do not think that this high-frequency relationship between the stock market and growth is due to the “wealth effect” of the stock market on consumer spending; our models suggest that wealth changes affect spending only gradually and over a several year period. Instead, one reason the stock market likely leads real GDP growth at this frequency is that investors trolling through company earnings statements and forward-looking guidance are likely to spot signs of improving prospects before the turn actually occurs in the data. In any event, the stock market seems to agree with our call that growth is turning higher. The main risk to our view comes from tax policy; income, dividend, and capital gains taxes are set to rise significantly at year-end unless Congress and the administration agree to extend them. We calculate that these tax increases would weigh on growth by roughly 0.8pp in 2011 relative to our base case, which assumes the administration’s plans to extend the tax cuts only for households making less than $250k per year passes. In contrast, if both sides agree on a full extension of the tax cuts, it would likely boost growth by roughly 0.2pp in 2011 relative to our base case. Indeed, the assumption that high-income tax rates will rise is the main reason we have slower growth slotted in for Q1 11 relative to Q4 10. We will likely make adjustments to our forecast as the tax policy negotiations play out in the coming weeks.
One argument often made is that tight credit conditions will keep the recovery from strengthening in coming quarters. We have argued in the past that it is not necessary for credit to be growing early in the recovery, as the initial pick up in consumer and business spending is instead driven mainly by improving labor income and profits. Indeed, both real consumer spending and business fixed investment have been growing solidly, even as consumer credit and bank lending to corporations have been contracting. As the recovery proceeds, banks typically start to ease lending standards, and, over time, this improvement in financing conditions helps to reinforce the recovery. The October Senior Loan Officer Survey from the Fed suggests that this process is occurring. A net 10.5% of banks reported easing standards on commercial and industrial (C&I) loans to large firms in October, while a net 7.1% eased standards to small firms; this was the second consecutive quarter that standards were eased for both types of firms. This is consistent with the idea that C&I lending by commercial banks will start to grow in the months ahead, and the more volatile weekly data suggest that loan growth may have turned into positive territory over the past few weeks. Similarly, a net 20.0% of banks reported an increased willingness to lend to consumers in October, the fourth consecutive quarterly rise. This suggests that consumer credit may start to grow in the coming quarters and indeed in September, consumer credit edged higher for the first time in 10 months. The bottom line is that the typical gradual move toward easier credit terms looks to be in place, and we do not think that credit conditions will prevent the recovery from proceeding.
Keefe, Bruyette & Woods FDIC Bank Failure Update: Three Failures Last Week.
With the number of FDIC bank closures totaling 140 in 2009 and our expectation for failures to accelerate throughout 2010, we are updating our weekly FDIC Bank Failure Update product to provide the relevant details of each failure. Last Friday, the FDIC announced 3 bank failures with $1 billion in total assets and an estimated cost to the FDIC of $204 million, or 20% of total assets.
Three Bank Failures- The FDIC announced 3 additional bank closings on Friday November 12, located in Georgia and Arizona. This weekend’s 3 failures brings the total number year-to-date to 146 and cycle-to-date to 314 institutions including 140, 25, and 3 in 2009, 2008 & 2007, respectively (Exhibit 1). In total, last weekend’s failed banks held $1 billion in assets and $922 million in deposits. The estimated losses to the FDIC totaled $204 million or 20% compared to the 2009 loss rate of 21% ($36.4 billion in losses on $171.9 billion in assets).
Concentration of Failed Assets & Deposits- Relative to the industry, last week’s failed banks had a higher concentration of construction loans at 20% of total loans, compared to an average of 7.2% for the industry. On the liability side this week’s failed banks relied more heavily on time deposits at 76% versus the overall industry at 47.7%.
Roll-Up Group- We continue to believe that our roll-up group, those banks with capable and willing management, sufficient capital, above average credit quality and regional opportunities, are likely to benefit by rolling up failed. Please see our July 6, 2009 note entitled "FDIC: Failures and Losses Mount, but Stress Creates Opportunity" and Exhibit 8 for additional detail on the complete group.
Performance of Roll-Up Group- In terms of performance the roll-up group has produced a cumulative return (since we introduced the group on July 6, 2009) of 8%, which is below the BKX’s return of 16% and the KRX’s return of 16%.
Potential Opportunities- In Exhibit 9 we attempt to map out potential opportunities for the roll-up group to gain market share by identifying banks, by region, with Texas ratios in excess of 100%. After incorporating this week’s failures, our potential opportunities table currently has 436 institutions with a Texas ratio over 100% with total assets of $230.8 billion. This compares to the FDIC “Problem List” of 829 banks with assets of $403 billion.
Location of Potential Opportunities- The majority of the assets we have identified in our potential opportunities list are located in the Southeast and Midwest with 29% and 22%, respectively, followed by the West with 18.5%. The Mid-Atlantic is not far behind offering 17.7% of total assets and the Southwest currently holds 11.7%. Additionally, the Northeast region presents the fewest opportunities with less than 0.8% of these assets.