the elusive bottom
Even market cheerleaders are struggling to find good news to rally around these days. With labor, capital, finance, real estate, and consumer markets all reeling from a half-decade of credit-fueled gluttony, and commodities markets cresting near all-time highs, it might seem a bit clichéd to highlight yet another bearish commentator — unless that bear is David Rosenberg, one of the few bulge-bracket economists to voice frequent and convincing skepticism about the “resilience” of modern capital markets and highlight the irrational optimism of the average investor. In his opinion, the markets have been in a secular downturn since the turn of the millenium, as the dot.com bubble morphed into the housing bubble, which evolved into the credit bubble, and which now seems destined to poison commodities prices, currency markets, trade flows, and even geopolitics through the end of the decade. Caveat investor…
Conference Call Notes, 14 August 2008
David A. Rosenberg, Chief North American Economist, Merrill Lynch
My sense is that we probably aren’t even past the halfway point yet of this recession, the credit losses or the house price deflation. Looking at whether equities may have bottomed or not on an intermediate basis, maybe the recent action to the negative side was an important inflection. In terms of what I do, which is trying to tie the macro into the markets, I have a very tough time believing that we have reached anything close to a fundamental low, either in the S&P 500 or in the long-bond yield, for that matter.
300-point rallies in the Dow happen in bear markets
We’re in a very confusing atmosphere. People didn’t really know what to make of a 300-point rally in the Dow the other day, but my main message was that 300-point rallies from the Dow don’t happen in bull markets. In fact, they never happened in the bull market from October ’02 to October ’07, but it has happened 6 times in this bear market and happened 12 times in the last bear market. You don’t get moves like that in bull markets. As Rich Bernstein has said time and again, “This is the hallmark of a recession and a hallmark of a bear market.”
How can there be recession with GDP still positive?
We are at a crossroad in the economy. The 2Q GDP numbers recently came in at plus 1.9%. The details of the number left a little to be desired, but it was still a positive number. Turn on CNBC, and everybody says, “How can there possibly be a recession with GDP positive?”
Employment has been down seven months in a row
The very next day we got nonfarm payrolls. It prints down 51,000 and frankly, it doesn’t matter whether it was below or above Wall Street expectations. The bottom line is that employment is down seven months in a row. In 60 years of sifting through the data here, that’s never happened before without the economy being in a classic recession.
GDP is useful but it has its limitations
I think the point that has to be made as an economist talking to a group of portfolio managers or FAs or investors, it is important to convey to clients that there is a lot of noise out there. GDP is useful, but it has its limitations. First, GDP is going to get revised. We thought we had a plus 0.6 in the fourth quarter; all of a sudden, it’s minus 0.2. Twenty percent of GDP is government. So, you really can’t fully concentrate on GDP when a fifth of it is state, local and federal government, unless you’re trading defense stocks.
You’ll miss a lot of action waiting for GDP to go negative
More to the point, if you’re waiting as an investor for GDP to actually turn negative, you’re going to miss a lot of action along the way. I think the best example is to just go back to Japan. They had a real estate bubble that turned bust and they had their own credit contraction back in the early 1990s. Guess what; Japan didn’t post its first back-to-back contraction of real GDP until the second half of 1993. By the time the back-to-back negative that people seem to be waiting for happened, the Nikkei had already plunged 50%, the 10-year JGB yield rallied 300 basis points, and the Bank of Japan had cut the overnight rate 500 basis points, which said a thing or two about the efficacy of using the traditional monetary policy response of cutting interest rates into a credit contraction (as we’re now finding out here in the US).
Dating the recession is a very scientific process
The point is we can’t make the assumption that we’ve avoided a recessionary condition in the economy, just because we have so far managed to avoid back-toback quarters of negative GDP. I’m just telling you as the economist that it is basically irrelevant. The only body that officially makes the call on the broad contours – when the recession started, when it ends, when the expansion starts, when it ends – is the National Bureau of Economic Research, the NBER. It’s a very scientific process. It’s not a gut check or a judgment call.
We should actually be welcoming the recession call
When they make the determination – it’s very interesting, by the way – when they make the announcement that the recession began, when they actually date it for us, traditionally we’re a month away from the recession actually ending. The announcement, in fact, is going to be a rather cathartic event, something we should actually welcome happening, but so far they are still taking their sweet time in making the proclamation.
Four factors used to determine recession
The NBER relies on four different variables. The first is employment. Now I’ve told you before; employment is down seven months in a row. Does employment go in the GDP? The answer is no. Is it correlated? Yes. Does it help grow the business cycle? Of course.
2) Industrial production
The next variable is industrial production. Does that go into GDP? The answer is no. Does it help grow the business cycle? The answer is yes. This is a number that comes from the Fed. The GDP comes from the Commerce Department. It’s a very important variable.
3) Real personal income net government transfers
The next variable, the third one, is real personal income excluding government transfers. This metric is now down four months in a row. Does personal income go into GDP? The answer is no; of course, it doesn’t. GDP is all about spending. Personal income goes into gross domestic income, which is another chart of the national accounts.
4) Real sales activity
The fourth variable and the only variable that actually feeds into GDP is real sales activity in manufacturing, retail and wholesale sectors.
Recession probably started in January
When I take a look at these four key indicators that define the broad contours of the business cycle, they all peaked and began to roll over sometime between October of last year and February of this year. I am convinced that when the NBER does make the final proclamation, it will tell us a that recession officially began in January. Of course, to any market person, this would make perfect sense, because of when the S&P 500 peaked. It did a double top into October, right when it usually does, before a recession begins.
This recession won’t end before mid-2009, in our view
Now I’m just giving you the rearview mirror. What’s most important to you folks is let’s look through the front window and see when this recession is going to end. The tea leaves that I’m reading at this point in time show that this recession is not ending any time before the mid part of 2009, which would mean that, if you’re looking for, not the Mary Ann Bartels intermediate bottoms, but the fundamental bottom, I don’t think you can expect to see it before February or March of next year, if I’m correct on when this recession ends. Historically the S&P 500 troughs four months before the economy actually hits its bottom point.
Profit as a share of GDP was at unheard of levels
The next question, of course, is what levels are we talking about? Again, I’m going to take what I do, which is earnings, and then talk about the appropriate multiple. What is the appropriate multiple at the low in a recession? In terms of earnings, I think that we have to understand where we’re coming from in this cycle. We’re coming from a situation where, because of all the leverage in the system, profits in the share of GDP went into this recession and bear market at 14% of GDP, which is unheard of. That’s never happened before. A lot of the reason why profits soared was because everybody turned to financials. There was this tremendous amount of leverage, and that accounted for half of just about everything in the cycle from GDP growth to employment to profits.
The profits share of GDP, again, as a proxy for margins, is now down to 12%. Think about that for a second. This terrible earnings recession so far has taken the share of profits from 14% down to 12%. The question is, if I’m right on a recession, where does the profit share of GDP go to at a recession trough? Well, consistently it goes to 7%.
We could get below $50 on operating earnings
Even the economists who are predicting a recession are going say, “Playing in a little recession, on average, troughs go down 25%.” The problem this time is that we have to overlay the revenue decline that actually comes from a recession with a much more significant margin, considering the levels from which we headed into this bear market and recession. So when I’m talking about that historically, what’s normal in a recession is that this profit share equals to 7% and we started at 14%, we are talking about a 25% decline in earnings. We can be talking about something closer to 50% peak to trough. The peak is $90 on a full-quarter trailing basis. It’s not beyond the realm of possibilities that we get below $50 in operating earnings. The first call consensus numbers is $105 earnings for next year. I give the odds of that happening at exactly 0.0%.
There is a good chance we test the 2002 lows
Now, I’m not at $50 for next year. We’re at $63 for operating EPS, but that means that the answer is no, I don’t feel that we’re too low on earnings. Usually you slap a historical trough multiple on in a recession. But typically, during a recession coupled with a credit crunch, the multiple bottoms at 12. You’re at a 12 multiple with $63 in earnings and you’re going to ask the question, “Are you talking about the possibility that we can actually test the … 2002 lows?” And the answer is that it is certainly not outside the realm of the possible. I’m not making that forecast, but what I am telling you is that there is a good chance that that could happen.
We are in a secular bear market
With that being respectful to the fact, I believe we’re in a secular bear market. I don’t even think that’s an opinion anymore. I think it’s a stylized fact. If you saw it, Rich Bernstein put out his performance asset mix table. Out of all the asset classes, stocks, cash, bonds, commodities, the only one to have a negative inflation-adjusted return over the past 10 years is the S&P 500. So I think we have to be honest about this. If it’s something like a 1929 and 1955 or 1966, 1982 type of secular bear market, I think this one actually started in 2000, it doesn’t mean that you don’t get cyclical bull markets along the way. We actually had a cyclical bull market in the context of a secular bear market that actually took the S&P to a new high. Of course, as I said before, half of that was unprecedented leverage, the stone process of unwinding.
I think that it is important now to recognize for our clients that we have a cyclical bear market being overlaid into a secular bear market. I think the message that we’re trying to send is that there is a different investing style and strategy for every part of the business cycle. One part of the business cycle is all about adding … data and risk to maximize your turns. Then there are times when it is all about preserving your capital and focusing on income, earnings, stability and dividend growth. I think that’s where we have been, and I firmly believe that’s where we will continue to be, at least over the course of the next 12 months.
Chapter 1 was the end of the res construction bubble
When I look at where we are in this book, and we continue to write chapters in this book and it is a book; this is an epic period. We are living through history. People will be writing about this in the future, no different than they wrote about the 1920s and the 1930s. Chapter one of the book was the end of the residential construction bubble, which I would tag as the first quarter of 2006, when housing started to peak and began to roll over at 2.3 million units. I continue to look back at that, 2.3 million units.
The natural level of demographic demand for housing in this country is annual demand of 1.45 million units. From 2003 until 2007, builders added on average nearly 2 million residential units per year, or 30% more, than the natural demand could absorb, because, of course, we were in a new paradigm. So the builders were building homes and condos as if we had the same demographics as the 1970s when the Boomers were buying their first refrigerator. This is a case of Global Crossing meeting D.R. Horton, and we are paying the price for that, even today.
Chapter two was the end of the home price bubble
Chapter two of the book was the end of the home-price bubble, and I would date that to the first quarter of 2007 when the Case-Shiller Index began to deflate year over year. Now, I want to make this point, and I want to make this point emphatically. Home prices in this country on average rose 20% per year for six years. That has never happened before. When you take a look at home prices in real terms, they’re still more than 30% higher today than they were when this mania morphed into a bubble back in 2001. So to those people who are thinking that we’re only 5% away from the low, I’d say I don’t think so. Make no mistake that there is going to be more deflation in home prices ahead – I think significant deflation – just as Freddie Mac put us on notice yesterday.
Chapter three was the end of the credit cycle
The third chapter was the end of the credit cycle, which, again, I would tag at exactly a year ago. I think the way we have to look at this, and we’re talking about how this affects our ability to navigate the portfolio and manage the macro forecast. This cycle saw the end of a 20-year secular credit expansion that went absolutely parabolic in the last 6 years and accounted for half the growth in just about every segment that’s forecast.
Chapter four was the end of the employment cycle
This is very big stuff and it’s taking on different forms. We have the end of the credit cycle as chapter three. Chapter four was the end of the employment cycle, which I discussed earlier, which started in December of 2007.
Chapter 5 is the first consumer recession since 1990-91
We’re heading into chapter five, and chapter five is the onset of the first consumer recession since 1990-91. I would argue this could end up being very similar to that six-quarter consumer recession that we endured from 1973-75. There are differences, but there are similarities. A lot of people like to compare this to 199091, because of the real estate flavor and the credit crunch, but there is actually a lot more going on that compares it to 1975.
I was around in the 1980s, and I remember that it played out very similarly. What people called resilience and people called contained and people called decoupling were all very pleasant euphemisms for lags. That’s what they are; they’re lags. There are built-in lags. Housing peaked in 1988, rolled over, the credit crunch intensified in 1989 when RTC got into real action. Then 1990 … two years after housing peaked, we had this very surprising consumer recession that caught even the Fed off guard.
The Four Horsemen
I wrote a report late last year titled The Four Horsemen. It was a regretful choice of words, because I kept on fielding questions as to whether or not I was, in fact, calling for the end of the world. I got to a point where my answer was “Just wait; it’s going to get worse than that.” In any event, who are the four horsemen? The four horsemen are credit contraction, deflation of both housing and equities, and that happened in the mid-1970s. Usually you’ll get one or the other. To have both housing and equities deflate on the household balance sheet, we’re talking about $30 trillion of assets. Half the assets on the household balance sheet are compressing dramatically right now. That last happened in the mid-1970s. We got credit contraction. We got deflation on the asset side of the household balance sheet that’s forcing the savings rate higher. We have employment, which I mentioned before.
Of course, food and energy – and, again, not just energy, but energy and food – and food is a bigger deal. Food is 15% of the household budget; energy is 10%. That’s a quarter of the household budget constrained by food and energy. Food is going to come down at a slower rate than energy will, but it’s already too late.
Oil prices are going down because demand is going down
People are saying to me all the time, “Gee, aren’t you going to turn more bullish with oil prices going down?” Well, oil prices are going down, because for the first time in this cycle it took $145 to break the back of the consumer. Quite amazing that it took that long, but it has happened. So we’re seeing true demand destruction in energy at a rate we haven’t seen in almost two decades.
It’s something to get an oil price decline that’s predicated on a new oil supply. I would keep that as a de facto exogenous tax cut; but when you’re getting oil price declines because of recessionary pressures cutting into energy demand, it’s no different than what happened in late 2000. That was the last time we had oil peel off as much as it is right now. I think it would have been a bit of a mistake for the economists at the end of 2000 to say, “Ah-ha, oil is coming down; I’m going to raise my 2001 GDP forecast.” You have to take a look at the reason why oil is going down, and the reason is not because of supply. The reason is because consumer demand is starting to go down. Again, the last time you had food and energy deviating so much from the long-run norm was in the mid-1970s.
Cash flow drain to household sector is $800 billion
When I take a look at the four horsemen and I try to come up with a number, the number I’m trying to come up with is a cash flow number. What is the cash flow drain on the household sector from the four horsemen in the coming year? The answer is $800 billion. So Uncle Sam, give me six more of those tax stimulus plans. That is a huge number. It’s equivalent to 12% of discretionary spending, which, by the way, is exactly the peak-to-trough decline in real consumer cyclical spending back in that 1973 to 1975 recession. The S&P 500 goes down peak to trough not by 20%, but more like 40%.
Three markers to turn us bullish
In terms of what are some of the markers that I’m weighing down to turn more bullish? I think this is very important. I look at not so much where am I going to be wrong, but looking at what are the things that will turn me more positive? There are three markers that I have laid down. The first marker is the personal savings rate. I have to see the personal savings rate go back to the pre-bubbles, normalized levels, which was 8%. I’m not talking about the Jurassic period here. I’m talking about where we were in the late 1980s and the early 1990s, before the last two bubbles. That’s why I said plural.
We had a tech stock bubble followed very quickly by a housing bubble. This had tremendous implications for perceived net worth and perceived future asset growth of the household sector. It had monumental impact on how people spent their after-tax income. That’s why we got to a point last year where briefly the savings rate got to negative for the first time since the 1920s. There was a belief system that we could retire on our assets, and now these assets are deflating and people’s expectations of how they’re going to retire is going to force that savings rate higher. That’s going to be very disinflationary, by the way.
I think it’s important to note that, in 2002, as the tech sector was deflating, Greenspan and Bernanke decided that it was a good idea to re-slate the housing stock as an antidote to the deflation in the tech capital stock. This is almost a piece of Mary Shelley’s Frankenstein; we built the monster, now we have to tear it down. I don’t know what else is left. We’ve had an equity bubble followed by a housing bubble, followed by a credit bubble. I don’t think there are any more rabbits in the hat to create the next bubble, unless that bubble is going to be in Treasuries, and maybe that is, in fact, going to happen. It’s pretty clear that the Fed is going to be concentrating a lot more in the future on non-traditional measures to ease monetary conditions, and not just cutting the Fed fund rate. Part of that may be reflating by expanding its balance sheet, which means that it’s not just talk. The Fed is actually going to add to its balance sheet, and that’s exactly what happened.
1) Need to see the savings rate go to eight percent
With the Bank of Japan and the operations they conducted back in the 1990s, this is just stuff to consider for the future. Let me just say that a savings rate of 8% would leave me feeling very good about the fact that we would have gone to a level of pent-up demand that would help us embark on the next bull market and economic expansion. That’s going to take quite a bit of time. This is a process. This a process we’re talking, even after the recession ends, that’s going to be an elongated recovery, as there was in the early 1990s, after that asset cycle. Remember, the recession might have ended in November 2001, but that did not give you a “get out of jail free” card as an equity investor, and certainly the recovery was a good two years away, even if the recession technically ended at the end of 2001. I’m talking about the markers that will turn me bullish for the next cycle. An eight percent savings rate, to me, would be a very critical launching pad.
2) Months supply below eight months
What else? Well, I doubt that anything is really going to bottom, including the financials, until we’re convinced that house prices have hit bottom. For that we have to look at the inventory to sales ratio, and there are different measures. There is the new inventory, which is a 10-month supply. There’s the resale; that’s 11-month supply. When I take a look at the Census Bureau data, which includes total vacant units for sale, single-family, condo, it’s more like 17-month supply. We need to include everything, including foreclosed properties. I have to see that number sliced in half. I have to see it down below eight months supply before I’ll be convinced home prices don’t bottom, at least the second derivatives start to turn positive. I have to see that metric at the eight-month supply. I’m keeping a very close eye on it. That will make me feel a lot more comfortable with turning bullish for the next cycle.
3) Interest coverage ratio has to come down to 10.5%
The third and last marker comes down to the household balance sheet. What I’m referring to here is interest coverage in the household sector. We have a record debt-income ratio, but that’s a stop-to-flow concept. I’m talking about interest coverage, how much are principal and interest payments from the record debt absorbing out of household income? It is 14.1%. It’s at a near-record high. We have never been in a recession with this metric at this level. So, that means there are too many things that are levels we’ve never seen before. The whole thing about economic bottling is you run the rest of it based on the past, and there are so many things that we’re entering into this thing that I’ve never seen before.
There is, I’d have to admit, a wide dispersion around the forecast I am providing. What I am really trying to do is put things into a certain perspective. What I know, being an economist, is that in some sense you’re a glorified historian. So when I take a look at the chart of interest coverage in the household sector, what do I see? I see that after the recession of the early 1980s, this interest coverage ratio got down to 10.5% by 1982 and, voila, that was the touch-off point for a multi-year bull market and economic expansion.
Then we had the recession of the early 1990s, and what do you know? In 1992, interest coverage went down to 10.5% again. That was the launching pad for a multi-year bull market and economic expansion. We’re 14.1% in this metric today. I know this historical record tells me that there is something about a 10.5% ratio that is a very cathartic event. The problem is that to get there from here would require the elimination of $2 trillion of household debt. So, maybe when NYU’s Nouriel Roubini talks about that the total losses could be up to $2 trillion, maybe he’s not talking through a paper bag.
Frugality is going to set in
As far as I know, there are only two ways to eliminate debt. You either walk away from it, which people obviously are doing, which is why we got these write-downs and these foreclosures, or you pay it down. I think people with a FICO score that they are concerned about are going to pay that down. That means that the savings rate is going to be forced higher. This, again, is going to be very, very disinflationary. It means that fashions are going to change. It means frugality is going to set in. We’re going to be living in smaller houses, driving smaller cars and living more frugally. It’s not going to be the end of the world; it’s going to be a necessary process to truly embark on getting the balance sheets down to more comfortable levels so that we can actually embark on the next cycle.
Intense deleveraging in the banking sector
The whole thing about being an economist is that you’re being requested to model behavior. What I found recently was three signs of significant changes in behavior. We obviously know of at least one investment bank that is taking aggressive action to sell assets and to deleverage. That’s going to force a lot of action in other parts of the industry. What we’re talking about here is intensified deleveraging in the banking sector.
Inventories cut by $62 billion despite tax stimulus
What else did we see? Well, those GDP numbers were just fascinating when you dig through them. Think about it for a second. How did businesses respond to the biggest tax stimulus of all time? They cut their inventory by $62 billion. Can you fathom that? Instead of boosting production as a result of the stimulus, they just allowed the stimulus to absorb past production. We already know that the inventory component went down another five points based on the July ISM number, so this inventory liquidation process is continuing.
Savings rate boosted despite stimulus too
Alan Greenspan cut his teeth on inventory investment cycles. So banks are deleveraging, and companies are liquidating inventories. How did households respond to the biggest tax stimulus of all time? They boosted their savings rate from 0.3% in the first quarter to 2.6% in the second quarter, which is only the third steepest increase in the savings rate in any given quarter in the past 55 years. Now you probably didn’t read that in the front page of The Wall Street Journal, but I find that to be a very relevant statistic.
So we have financial sector deleveraging. We have business sector inventory liquidation overlaid with the households boosting their savings rate. These are new themes, and the theme is about getting small. That’s going to play very well into Rich Bernstein’s decision two months ago to allocate an extra 15 percentage points to his fixed income portfolio. Now we’re talking about fixed income. We’re talking about bonds that are high quality and have non-callable protection.
Nominal GDP growth has highest correlation with yields
I’ll tell you that the really key forecast next year coming from the economics department here is the nominal GDP, nominal, price times quantity, because we’re calling for nominal GDP growth next year to average 1.5%. That is going to be very bullish for sectors that have proven earnings stability and reliable dividend growth, and it’s going to be very bullish for bonds. I say that, because I know that the critical driving factor for bonds is not fiscal deficits. It’s not the dollar and, guess what, it’s not commodities. Nominal GDP growth has the highest correlation. People look and they say, “Four percent 10-year note; who’d want to touch it?” The reality is that nominal GDP growth this year is averaging 4%. The fact that the 10-year note is averaging 4% is not really a big mystery, if you’re looking at the macro underpinnings.
Now, if I’m right on 1.5% nominal GDP growth for next year, all I can tell you is that the last time we had a condition like that was in 1958. All I can tell you is that 1958, the funds rate averaged to 1.5% and the 10-year note averaged 3%. If you’re going to ask me if we have a realistic chance of going back and retesting the June 2003 lows and the 10-year note or the March 2008 lows and the 10-year note, I firmly believe that’s going to happen. I believe that’s going to also provide you with very handsome total returns.