The Recession Buy Indicator

Stock_market_logicInteresting piece by Mark Hulbert in the Sunday NYT on the Recession Buy Indicator. This timing signal was created by Norman Fosback author in the mid-1970s of the popular investment textbook “Stock Market Logic.” The book, though out of print, was first published in 1976, and has gone through 18 printings, selling more than a half-million copies (used copies are available at Amazon and elsewhere).

The indicator goes something like this:

"The [ Recession Buy Indicator] focuses on the four business barometers that together make up the federal government’s index of coincident economic indicators. These four focus on industrial production, manufacturing & trade sales, nonfarm payrolls and personal income. The Recession Buy Indicator is triggered when — as is the case today — each of these four gauges is below its level of six months earlier. On such occasions, Mr. Fosback considers the economy to be in a recession or very close to it."

I love this sort of approach, for several reasons: It is a contrarian play, which has tremendous appeal to me; it provides an objective, mathematical way to make a buy or sell decision; it has a significant body of backtesting.

Since creating the indicator in 1979, it has triggered 4 buy
signals. Over the 12 months following, NYSE average gains were 37%, and
after 3 years, 106% (triple the stock market’s average).

Before you mortgage the house to buy long dated index calls, however, there are a few caveats worth noting. The previous signal was in February 2001 — 18 months before 2000-2 bear market lows. Buyers of that signal got crushed in the ensuing selloff.

That sort of miss is the reason Ned Davis Research is less of a fan. While the average performance of the indicator was very good, the misses are enough to caution not blindly following the signal. And, during some recessions, the stock market’s actual bottom came much later.

"That is one reason, he said, that his firm doesn’t mechanically issue a buy signal six months after the economy begins to turn downward. Instead, it prefers to await confirmation from a number of its other indicators that a bottom has been formed. In the current market, that confirmation has not yet come, he said, and his firm has a policy of not trying to predict when it will."

Hulbert’s use of data and statistics is why he is always an interesting read . . . Good stuff.


UPDATE: May 18, 2008 10:23am

Alternate views can be seen here and here .



An Alarm Is Blaring: Time to Buy
Mark Hulbert
NYT, May 18, 2008

Stock Market Logic a Sophisticated Approach to Profits on Wall
Norman Fosback
Hardcover: 384 pages
Publisher: Inst for Econometric Research (1976)

Print Friendly, PDF & Email

What's been said:

Discussions found on the web:
  1. RB commented on May 18

    What do you think of the latest Time magazine cover (“Surviving the Lean economy”?

  2. rww commented on May 18

    Four “buy” signals, one of which was disastrously wrong? That’s a reasoned basis to buy stocks? This would be a fine example of confusing pattern and causation except that four events do not even form a pattern. I don’t understand why anyone reads Hulbert.

  3. Philippe commented on May 18

    Why the “Great” Depression? or the cost of avoiding normal economic cycles.

    “1931-36: Police stand guard outside the entrance to New York’s closed World Exchange Bank, March 20, 1931. (copyright unknown)
    Economists are still divided about what caused the Great Depression, and what turned a relatively mild downturn into a decade long nightmare. Hoover himself emphasized the dislocations brought on by World War I, the rickety structure of American banking, excessive stock speculation and Congress’ refusal to act on many of his proposals. The president’s critics argued that in approving the Smoot-Hawley Tariff in the spring of 1930, he unintentionally raised barriers around U.S. products, worsened the plight of debtor nations and set off a round of retaliatory measures that crippled global trade
    All rules have their exception (s) if one would look at the US business cycle from 1930 until 1940 he may assess the price of jumping too soon, and when looking at the European business cycle (except UK) during the same time scale he may assess the price of staying too late.
    Europe went into a prolonged crisis well after the USA and the UK. How would these cycles be responding through a more flexible monetary policy, my guess is a prolonged anaemic growth?

  4. Will Rahal commented on May 18

    Last night I actually posted charts on all of these indicators!

    If we had had a more significant bear market, Fosback’s indicator would be screaming for a buy.

    It is unlikely, however, that at this point the financial excesses of this cycle(and last 25 years) have been corrected.

    A more likely scenario would be the lowering of earnings expectations followed by a new leg down in the Stock Market.

  5. Steve Barry commented on May 18

    Hulbert is an interesting read…unfortunately, I have read almost every one of his columns for the last 6 months and he is trying so hard to make every one have a bullish conclusion for some reason. He wasn’t like that in the past. Here are some bearish angles that need to be at least asked:

    1) How does $125 oil affect the consumer? I just yesterday had my tank run to near zero and paid $76 to fill it.

    2) How will housing affect the consumer? Notice it’s declining for one of the few times in US history.

    3) Based on peak earnings, the market is quite expensive (read Hussman)

    4) The Shanghai market has plunged 50%…what does this say about the Chinese economy? Or do markets only predict in the US?

    5) One word…inflation

    6) Hulbert is such a fan of contrarianism…anyone check out the fresh high on the Citigroup Panic/Euphoria model? it just printed plus fifty freakin’ four!!!! I thought +.39 was a record. Take that, jack. Euphoria in the face of poor fundamentals could lead to a non-linear event.

  6. Jim Haygood commented on May 18

    Trading system developers quickly learn that the period from the 1950s through the 1970s was a kind of golden age for system trading. Recessions occurred with a near-perfect four year periodicity. Each one was signaled by rising short-term rates, and lasted a year or so. Then plummeting short-term rates signaled the advent of the next bull market. The neophyte developer is moved to exclaim, “Man! Is this EASY, or what?”

    Unfortunately, the mold broke after 1982. Suddenly, the four-year periodicity of recessions disappeared; economic expansions became decade-long affairs. Another no-show was the “average” bear market of 12 months or so: for the 18 years from 1982 to 2000, the rarely-seen bear market (1987, 1990) was only a single quarter (three months) in length. The upshot is that models which tested so beautifully during the 50s-70s “golden age” — including many developed by Fosback, Zweig, Ned Davis, et al — either worked less well after 1982, or even fell apart.

    The 2000-2002 bear market lasted for an unusually long 31 months, so models based on falling interest rates or economic series (which imnplicitly assumed a typical 12-month bear market) gave disastrously early signals.

    This time around, in 2007-8, the consensus wisdom would have us believe that the near-bear market during the near-recession lasted a mere 6 months, from Oct. 2007 to March 2008. If that turns out to be true, then the recent action would be almost an analogue of the LTCM crisis in 1998, when a brief financial crisis pulled stocks down sharply from a record high. But then aggressive Fed easing put the bull market back on track for fresh record highs, printing a classic “V-bottom.”

    Some important differences exist, though. For one thing, commodity prices were at a low point in 1998, not running away as they are now. For another, the US dollar index was around 100, instead of the low 70s today. Thus, inflation was not nearly as much a concern in 1998 as it is today. And perhaps most important of all, the overhang of questionable and nonperforming debt was far smaller than now.

    On the economic front, 1998 was one of the highest-growth years for adoption of web-based sales channels. No comparable paradigm-changing technology is undergoing mass adoption today.

    For all of these reasons, I don’t think the 1998 analogue is going to hold; 2009 is not going to be a “party like it’s 1999” year. Greenspan’s conundrum — that is, why Treasury buyers are willing to accept negative real rates of return — is likely to be resolved in 2009 by sharply rising nominal yields. And that’s not going to be good for stocks. Hell, not even for real estate. Sounds like a train wreck, in fact.

  7. Marika commented on May 18

    Strong personal income growth was the reason for NDR model not screaming buy. In addition, NDR model is calibrated for a recession but we are not in a recession (2008 is very similar to 1998 the end of the world panic, even bulls as Cramer panicked and sold in 1998). Often, if you wait for a perfect buy signal, you will often miss the perfect bottom.

    Record low AAII bullish sentiment and NYT declaring a recession on their front page in March was enough confirmation for me to load up the truck.

  8. Marika commented on May 18

    Will Rahall,

    Why are you using Non-Farm Payroll excluding Health and Education? NBER does not exclude health and education jobs.

    In addition, change in industrial production and incomes are not at the recessionary levels of 1991 and 2001.

    According to the data (that NBER considers) we are not in a recession.


    Steve Barry,

    Loading up the truck in March has been very profitable. I am up by 40% so far.

  9. Scot commented on May 18

    40% on a 10% move? How much leverage are you using?

    Geez, thats way too risky for my blood.

  10. Mind commented on May 18

    Record deficits. Everything’s made in China. US = military state. Peak oil. The housing equity ATM is closed. Record foreclosures and bankruptcies and still going up. Zero household savings rate. Ethanol = world food shortage. Dollar worth s**t, and so on. This is the negative storyline.

    Under these conditions, not easy to have confidence in a recovery according to a “typical” historical scenario?

  11. Winston Munn commented on May 18

    I am not nearly so optimistic that the bottom is in – it is different this time because it is different every time, and past is not prologue but a rough first draft.

    Keep in mind that this has been the slow motion meltdown, and it continues.

    For example, Center for Budget and Policy Priorities reports:

    “Thirteen states, including several of the nation’s largest, face a combined budget shortfall of at least $23 billion for fiscal 2009. Another 11 states expect budget problems next year or the year after. The initial reports for 2009, which runs from June 2008 to June 2009 for most states, suggest states are returning to a time of budget deficits.

    This brings the total number of states identified as facing budget gaps to 24 — close to half of all states.”

    The impact of these deficits has yet to be felt as lowered state services and higher taxes and higher unemployment.

    If there is a slight rebound, it will only point to the shape of a W recession.

    Things are only now beginning to get interesting.

  12. DownSouth commented on May 18

    “…it is different this time because it is different every time…”

    “Keep in mind that this has been the slow motion meltdown, and it continues.”

    Thoughtful comments, Winston Munn.

    I have come to believe that what we are seeing today looks just as much like the 1920’s as it does the 1930’s, and that Bernanke is wrong in comparing today’s troubles to those of the 1930’s. We’ve seen some economic turmoil, but the dominant attitude is still optimism. As Frederick Lewis Allen observed of the unshakable optimism of the 1920’s: “The lesson was plain: the public simply would not be shaken out of the market by anything short of a major disaster.”

    There was a huge real estate bust in the 1920s. “Thousands of country banks, saddled with mortgages and loans in default, ultimately went to the wall. In one of the great agricultural states, the average earnings of all the national and state banks during the years 1925-1929, a time of great prosperity for the country at large, were less than 1-1/2 per cent; and in seven states of the country, between 40 and 50 percent of the banks which had been in business prior to 1920 had failed before 1929.”(Frederick Lewis Allen) And main street began to falter in 1927 and 1928. But despite all these telltale signs, the stock market continued its inexorable rise.

    Also looking at the curve that Steve Barry links to, we see “slow motion” you refer to. Stocks trade at the top of the trading range for years–often for more than a decade–the fall to the bottom normally takes years, where stock prices remain for years.

    Also, what do current P/E’s look like? Are they not running significantly above the historical average of 10?

  13. Eric commented on May 18

    This article would have been a quite a bit more impressive if it had appeared in early March. Telling us to buy “when everything seems gloomy” is pretty silly with the S&P 500 down all of 2% YTD. The article is not praiseworthy.

  14. SteveC commented on May 18

    One of the issues I have with backtesting, is a) how far back you go, b) what variables are you going to include. For many of these systems, the sample is not long enough or includes the wrong variables to be significant.

  15. Steve Barry commented on May 18

    Just saw an amazing give and take with Maria Bartiromo and Mohemmed El-Arian from Pimco on the WSJ Report…I’ll paraphrase:

    MB: Are you still worried about the economy slowing or inflation?
    ME: I am worried about the economy slowing and also worried about inflation levels. What we have avoided is a complete breakdown of the financial system.
    MB: OK, so we don’t have to worry about a breakdown in the financial system now, so what is your outlook?

    Amazing bit of deduction by Maria isn’t it…Mohammed went right along with her. We are driving in a truck, the breaks just failed, we just missed a cliff, so the problem is over?

  16. Steve Barry commented on May 18

    Even more amazing..after 20 years, McLaughlin Group is on WCBS in NYC???? I’m in shock…didn’t even know it.

  17. Winston Munn commented on May 18


    There is reason for lack of optimism. The ARM reset problem has not disappeared while it will still take years to work through unsold inventory, the overbuilding of previously financed commercial real estate continues while demand for retail space falls, banks continue to reduce their balance sheets and tighten lending standards, the ABCP market is at all time lows, tax receipts keep falling (California went from a $10B budget shortfall in Nov to the most recent estimate of $17B.), and jobs are disappearing almost as fast as consumer sentiment is falling.

    These kinds of problems compound as one feeds off another – the effects are insidious and inexorable – like trying to walk uphill with an ever-expanding weighted backpack.

  18. bluestatedon commented on May 18

    “We’ve seen some economic turmoil, but the dominant attitude is still optimism.”

    Optimism on whose part? Certainly that’s not true of large segments of the country, across a wide range of demographic groups. It’s also not borne out by the most recent UM Survey of consumer attitudes, which are the most pessimistic they’ve been for many years.

  19. Mind commented on May 18

    Optimism by the investing class – those with $$$$ thirsting for an opportunity to make > 1.5% on it.

  20. DownSouth commented on May 18

    ☺”Optimism on whose part?”
    Posted by: bluestatedon | May 18, 2008 1:04:04 PM

    You are absolutely right, bluestatedon. Consumer confidence and other surveys that ask questions like “are we headed in the right direction” or inquire about people’s sense of financial wellbeing indicate a malaise not seen in decades.

    However, I would still argue that, by historical standards, stocks are still very expensive. And that is what I refer to when I speak of “optimism.”

    Again, look at the graph Steve Barry linked to and are stocks on the average not trading at P/E’s over 15, well above the historical average of 10?

    If stocks were trading at a P/E of 10, wouldn’t the DOW be at about 8000, instead of its current 13000?

    And if we look at Steve Barry’s graph, wouldn’t we be looking at a 7000 to 8000 DOW if stocks were trading towards the center of their trading range?

  21. Tom F. commented on May 18

    “Before you mortgage the house to buy long dated index calls, however, there are a few caveats worth noting. The previous signal was in February 2001 — 18 months before 2000-2 bear market lows. Buyers of that signal got crushed in the ensuing selloff.”

    Yeah, Don Hays can tell you all about that. He is definitely one who got a little too anxious to buy and got his dick knocked into the dirt as a result.

  22. jagmohan Swain commented on May 18

    Hulbert articles never fails to draw criticism. All one needs to do is to read the comments for his columns at Marketwatch. If you ever needed anecdotal evidence of pervasive negativity surrounding this business cycle this is it. People say that most recessions one could see miles off. Is that so? Then why nobody saw the recession that was coming in 2001. As per WSJ 90% of economists were bullish on economy for year 2001. Recession started in March 2001-Nov 2001.

    This again underscores the point that the prevailing mood among investors, business managers and people in general has an impact on the depth of contraction. If people have braced themselves for output contraction the resulting adjustment in production output would be much lesser compared to say if every one is expecting a happy time going forward. It’s the magnitude of adjustment required in production output that determines the employment situation.

    Fortunately all pervasive gloom has done the main street a great service by keeping the economy at peak efficiency. I remain bullish on the market, short corrections aside.

  23. Will Rahal commented on May 18

    During the last 2 recessions, Health & Education Employment actually went up.
    What is interesting about looking at the Non-Farm-Payroll ex HE is that it is the yr/yr % change is negative. This level has always been associated with recessions.

  24. DL commented on May 18

    Hard to imagine that we are now seeing the worst of the recession. We have yet to see the BEA admit that GDP has turned negative. Unemployment is only 5%. At the same time, investor sentiment is pretty optimistic given that the R2000 index has been declining for ten months, the price of oil has risen 400% over the last 5 years, and the inventory of unsold homes is at an 11 month supply.
    The “buy and hold” approach is unlikely to work well over the next 12 months (except perhaps for a few specific sectors).

  25. Kurt commented on May 18

    Hey Winston,
    That 17B for CA wasn’t good enough for us, so we recently upped it to 20B. But it’s okay, were going to borrow a few dollars from the Lotto!
    I am using the big up days to add to LEAP put positions in the consumer discretionaries.

Posted Under