|
|
|
|
|
|
February 5, 2010
Mike Farrell will be appearing as a guest on Consuelo Mack’s “WealthTrack” this weekend. (Dates and times will differ by location, so check your local listings. Video replay is also available on the WealthTrack website.) On the show, Mike discusses a number of macro themes, and several of the graphs maintained by our research group will be highlighted. Since graphs are rarely on screen long enough for thoughtful contemplation, we reproduce them here along with some context.
The first two graphs illustrate how different actors in our economy are behaving through the current environment. Households, businesses and financial institutions alike have begun the process of increasing their savings, refinancing their debt and de-levering their balance sheets. It’s a painful but necessary process. The progress made by the private sector, however, has been more than offset by the borrowing and spending of federal, state and local governments. The headline savings rate that we usually think about is the household one, but the same calculation can be applied to the government. On a national level, think of households, businesses, financial institutions and governmental entities all contributing their savings to one big piggy bank. The first graph below shows what each is contributing to our national piggy bank in the form of net savings. Gross savings is simply income less expenditures, and net savings is calculated by subtracting the consumption of fixed capital from gross savings. As the graph shows, net savings of the private sector has been generally rising and net federal government savings lately has been plummeting.
The graph below sums up the total of private and government net savings and presents it as a percentage of gross national income. It’s clear that government spending is draining the national piggy bank.
The third of our graphs that appears on this weekend’s WealthTrack is the one below, which demonstrates how the mortgage market-and thus the housing market-is on government life support. There has been a collapse of non-conforming loan originations or bank-retained originations, and virtually no mortgage-backed securities have been issued in the last two years without the benefit of a government wrap of one kind or another.
Posted in Consumer, Current Events, Government Policy, Mortgage Markets | No Comments »
February 2, 2010
On many occasions and in many different forums, The Federal Reserve has expressed every intention of winding up its $1.25 trillion program of buying agency MBS by March 31. “To provide support to mortgage lending and housing markets and to improve overall conditions in private credit markets,” the Fed explained in the January 27 FOMC statement, “the Federal Reserve is in the process of purchasing $1.25 trillion of agency mortgage-backed securities…. In order to promote a smooth transition in markets, the Committee is gradually slowing the pace of these purchases, and it anticipates that these transactions will be executed by the end of the first quarter.”
By now, everyone is taking the Fed at its word. The budget document submitted by President Obama yesterday regarded it as a foregone conclusion. But the budget also included this section: “This MBS purchase program is widely credited with pushing down mortgage interest rates, which according to [Freddie Mac] reached an all time low of 4.71 for the average 30-year fixed-rate the week ending December 3, 2009.” Indeed, the spread between the 30-year commitment rate and the 10-year Treasury has narrowed from 260 basis points on January 7, 2009 to 133 basis points yesterday. But over that time the mortgage commitment rate has traded in a relatively narrow band and fallen from 5.10% to 4.98% while the 10-year Treasury yield has risen from 2.50% to 3.65%. The graph below shows the 30-year mortgage commitment rate versus cumulative Fed purchases since the program started in the beginning of 2009. So while it may be a stretch to suggest that mortgage rates have been “pushed down,” it is fair to ask how much higher they might be without the government intervention.
Not only is it a fair question, it is the question of the moment. Come March 31, the Agency MBS market is going to find out how it will perform without the government’s involvement (the Treasury has already quit its buying program). The Fed has already begun to wind down its weekly buying. For what it’s worth, the last several weeks the Fed has cut its buying pace in half, averaging about $12 billion in net purchases per week, down from about a $25 billion weekly average last summer and fall, and prices have been steady to somewhat higher.
The table below sets forth where the Fed has been buying in the mortgage coupon stack. Over 80% of the $1.16 trillion purchased to date has been in 5% coupons or below, with the majority (43.2% or $501.5 billion) in 4.5s. Interestingly, since the beginning of the buying program there has been little significant difference in price performance by coupon despite the Fed’s concentration in the lower coupons. As far as predictions go, our guess is no better than anyone else’s. But the evidence (particularly the strength in the higher coupons) seems to suggest that the market for Agency MBS is deep and liquid and waiting to see what happens when the Fed steps aside.
Posted in Current Events, Government Policy, Mortgage Markets | No Comments »
January 29, 2010
It is very hard to presume any sense of accuracy when answering the question, “What will market conditions be in the future?” There is the rub, because the future is what matters most to an investor. Investors, however, know the past and present cold. Thus, it is somewhat more comforting to ask the same question in this way: “How long will current conditions persist?”
The Federal Reserve understands this problem, and nodded toward it yet again in their January 27 FOMC statement by maintaining their commitment to keep the federal funds “exceptionally low…for an extended period.” Presumably, this means the current rate at 0% to 0.25%. That still leaves some uncertainty on the table, but it provides a semblance of forecastable conditions for one point on the government yield curve. What about the rest of the government curve? And the curve for other asset classes? There is no such guidance from the Fed, because the fed funds rate is all it controls (please, no angry blog comments about the market manipulation resulting from the Fed’s active involvement in the financing markets and open market purchase operations).
For answers to those unanswerable questions, we have to fall back on the past and present. The two stacked graphs below try to show the relative yield at two points on the government yield curve (Fed funds and the 10 year Treasury) and the mortgage finance curve (the current coupon mortgage-backed security and 3-month LIBOR). The vertical red lines indicate the beginning of a Fed easing cycle. After the Fed started easing in June 1989, the government curve steepened to more than 300 basis points and stayed that way through 1995, six years later. After the January 2001 easing campaign got underway, the government curve stayed steep for about five years. The mortgage finance curve is similar (although it never inverted). Today, we are a little over two years into the easing campaign. If history is a guide-and there is, of course, no guarantee that it will be-these curves will likely remain steep for some time.
 Source: Bloomberg and Federal Reserve  Source: Bloomberg and Federal Reserve
Posted in Bond Markets, Monetary Policy | No Comments »
January 26, 2010
The past two days have been disappointing ones for housing market followers. Yesterday’s existing home sales showed a much larger than expected drop from the previous month (from 6.54 million in November to 5.45 million in December on a seasonally adjusted annual basis), and today’s S&P/Case-Shiller home price indices came in below estimates. Also reported yesterday, to much less fanfare (it’s not even on the Bloomberg economic release calendar!) was the Federal Reserve’s preferred measure of national home prices: the LoanPerformance National Home Price Indices. The Fed uses this index to calculate the value of real estate assets for the Flow of Funds report released every quarter. Let’s just drop all of these data points onto a graph, shake vigorously, and see what comes out.
Both home price indices have been normalized to 100 in January 2000, and their most recent data points are for November 2009. Existing home sales data is for December 2009.
The first thing to note is the disagreement in November between Case-Shiller and LoanPerformance, the former showing a small rise in home prices and the later showing a small decline. The two indices use similar methodologies, and tend to move together over longer time periods. We won’t hazard a guess as to the reason for this single-period divergence.
The second, and most obvious, feature of the graph is the Cash-For-Clunkers-style heartbeat pattern in existing home sales. The original cut-off for the first-time homebuyer credit was November 30, 2009. Congress approved the extension on November 5, 2009, but most homebuyers who wanted to use the credit made their purchases in September to make sure that they closed before the deadline. Therefore, we can assume that the peak in tax credit-induced buying was November, which as we see briefly spiked back to 2006 buying levels. The dramatic fall in December closings represent sluggish activity in October and November, months in which we stated previously would likely represent levels of “activity that we would call ‘normal’, unmolested by incentives”. It’s clear now that there was nothing organic about the pick-up in sales activity or prices in the back half of 2009. We look forward to seeing how S&P/Case-Shiller and LoanPerformance home price indices respond to this lower level of demand when we receive December data.
Posted in Government Policy, Housing | No Comments »
January 22, 2010
“Administrative” issues were blamed for yesterday’s disappointing weekly jobs numbers, which showed a 36,000 rise in initial claims to 482,000. Continuing claims continued its steadily downward trend (on a seasonally adjusted basis, of course). In past recessions, a continuous fall in continuing claims was a reliable indicator of the beginning of a new expansion. We aren’t saying it’s different this time, just that the data set has changed. As we’ve mentioned before, the duration of unemployment is significantly longer than it’s ever been (since we started measuring), reaching an average of 29.1 weeks in December. Considering that basic unemployment insurance only runs for 26 weeks, it’s not surprising to see continuing claims falling as more people are expiring these benefits before finding a new job. A new category of claims was created in June of 2008, when the Emergency Unemployment Compensation 2008 program was established. It has since been adjusted several times: expanded on 11/21/08, extended on 2/17/09, and expanded again on 11/6/09. Most headlines still focus primarily on initial and continuing claims, but plenty of observers have been watching emergency claims, noting its growth as unemployed workers roll into this new category. It appears that it won’t be long before there are more emergency claimants than continuing claimants. Since emergency claims are only given to us in non-seasonally adjusted (NSA) form, that is the way we show both kinds of claims below. Continuing claims follows an obvious seasonal pattern, but emergency claims marches to its own drum.
And because we know you’re curious, we’ve provided the pretty stacked area graph for you, also in NSA form.
The current iteration of the emergency benefits program provides for up to 73 weeks of unemployment insurance for certain claimants, depending on when they file for benefits. We are now over 100 weeks into the recession, if it’s still officially going, of course.
Posted in Current Events, Labor Markets | No Comments »
January 19, 2010
How should the economic success of a country be measured? On what it consumes or on what it earns? Peter Schiff of Euro Pacific Capital says it’s the latter, and we have to agree. “Using GDP as the main financial indicator is equivalent to judging a man’s success by the cost of his house, car and wristwatch…. [T]hese figures merely indicate a level of spending and have nothing to do with earning power.”
If not GDP, then perhaps tax receipts, and by this measure the United States is far from recovery. According to the Treasury Department’s latest Monthly Treasury Statement , released last week, receipts are falling off a cliff, leading to deeper and deeper deficits. These cash flows are reported on a monthly basis and are pretty lumpy, so we present them on a trailing twelve month (LTM) basis.
On a trailing twelve month basis, calculated monthly, from the table below it is clear to see that tax receipts are essentially back to where they were in calendar 2004, while outlays have ballooned by almost $1.2 trillion. Even if receipts were to recover to the high point of 2007 (a highly unlikely outcome in the near term), the deficit would still be almost a trillion dollars.
The graph below (inspired by Paul Kasriel of Northern Trust) paints the same picture but a little differently. The year-over-year percentage change in LTM receipts has been dropping at a double digit rate since April 2009, even as the year-over-year growth in outlays has actually been moderating.
Some would look at these graphs and table and come to the conclusion that a higher tax rate is in order. We look at these data and come to a different conclusion. First, encourage, support and stimulate those activities that result in taxable outcomes. In fiscal 2009, individual income taxes totaled $915.3 billion and accounted for 43.4% of total federal receipts; corporate income taxes totaled $138 billion and accounted for 6.6% of total receipts. In fiscal 2008, individual income taxes totaled $1,146 billion and corporate income taxes totaled $304.3 billion. In other words, individual and corporate income taxes fell almost $400 billion from 2008 to 2009. On January 7, the Rockefeller Institute released its quarterly state revenue report which tells the same tale. “Both nominal and inflation-adjusted figures indicate that the first three quarters of 2009 marked the largest decline in state tax collections at least since 1963,” the report said, with across the board weakness in personal, corporate, property and sales tax receipts. Second, obstacles to the structural health of federal, state and local governments budgets are massive and will require huge spending cuts. The increases in federal spending from fiscal 2008 (which ended September 2008) to fiscal 2009 (which ended September 2009) were largely in defense spending (+$42 billion), Health and Human Services (+$96 billion) and Treasury (+$222 billion excluding changes in debt service costs). States have already started across the board budget cuts, the federal government must do the same.
To us, a rebound in GDP only reflects a rebound in consumption, and today’s consumption is fueled to a large extent by growth in government spending, incentives and, most significantly, borrowing. A rebound in tax receipts, sans tax increases that stymie economic activity, would reflect growth in our country’s earning power.
Posted in Consumer, Government Policy, Macro Economics | 1 Comment »
January 15, 2010
As of January 2010, the duration of the current recession stands at 25 months and counting. By lasting this long, it has vaulted into 5th place on the career list, surpassing the 24-month long recession of January 1910 to January 1912. It’s the Ken Griffey of recessions. Next one in our sights is the 32-month long post-Civil War recession that ran from April 1865 to December 1867.
For all you statistical purists out there, please note that the list kept by the National Bureau of Economic Research, the official arbiter of recessions, starts in 1854, and therefore does not tally some earlier American recessions, like the Panic of 1797, the Depression of 1807, the Panic of 1819 and the Panic of 1837.
Of course, the NBER declares the beginning and end of recessions retroactively (it declared the official December 2007 start date of the current recession in November 2008), so it may turn out that our current recession may already be over. Is it? Economist Robert Hall, who heads the NBER’s Dating Committee, suggested in a December 4, 2009, interview with Bloomberg that the trough in job losses may be the signal that the recession is over. On the other hand, Martin Feldstein, a member of the Dating Committee and the former president of the NBER, opined later that month that the recession isn’t over. Whether it is a recession or not, Feldstein said that growth in 2010 will be weak due to increased savings and lower spending. “Thrift in the long run is a very good thing, but increasing thrift as you come out of a recession is going to be a drag.”
Recall that the NBER defines a recession as a “significant decline in economic activity spread across the economy, lasting more than a few months, normally visible in real GDP, real income, employment, industrial production, and wholesale-retail sales.” Based on these variables, it’s a mixed bag. Industrial production looks to be turning, but it’s still below 2003 levels. Personal income growth is sluggish. Retail sales ex-autos and gas, according to the Commerce Department release yesterday, fell 0.3% in December after a strong November gain of 1.0%. The jobs picture is looking intractable. The number of people who are unemployed and who are receiving unemployment benefits (”continuing claims”) is declining on a seasonally adjusted basis-down 2.3 million from a cyclical peak of 6.9 million in June 2009 to 4.6 million at last report-but the number of people claiming emergency unemployment compensation after their standard benefits have run out has risen by 2.4 million to over 5.0 million over the same time period. (Hat tip to David Ader of CRT Capital Group.)
 Source: Bloomberg William Hester at Hussman Funds looked at the primary NBER components and concluded that it is too early to tell. “With real GDP still down more than 3 percent from its peak,” he writes, “it would go against precedent for the group to declare the end of a recession with the mixed signals that are currently in place.”
The decision-making process of the NBER is slightly inscrutable. But from where we sit, our current recession is not ready to be called over, making it a shoo-in for the Recession Hall of Fame.
Posted in Macro Economics, Monetary Policy | 1 Comment »
January 12, 2010
There have been a lot of “lost decade” charts put out since the new year, most either focusing on the stock market or job growth (both being effectively zero or worse). There have been very few jobless decades, but in truth we just experienced one: non-farm payrolls were 130.5 million on 12/31/99 versus 130.9 million on 12/31/09. After the slew of Bureau of Labor Statistics data in last Friday’s jobs report, we saw the chart below show up in a lot of places.
The employment rate is the total number of employed persons divided by the total civilian non-institutional population. A good many versions of this chart started in the mid-1980’s, which showed the metric making new lows, but the BLS has been measuring this since 1948. As we looked at the chart we started asking the reasonable question: why is a smaller portion of the work-age population, just 58.2% at latest count, actually working? But then we asked another question: Why was such a large portion of the work-age population working in the last 20 to 25 years? The employment numbers are broken down into age brackets, and many observers have pointed out that the younger portion of the population is taking much of the hit during this downturn. As we looked at the various age buckets, we recognized that there are always cyclical and structural forces affecting employment. The cyclical effects are typically recessions, when total employment falls. But when you break down employment by age group, you realize that there has been a rather large structural effect going on for the last, say, 45 to 63 years: the Baby Boomers.
The following chart looks at employment broken into buckets, and is a good visualization of the “pig in the python.”
Notice first that the only age bracket being measured on the right hand scale is 20 - 24 years old (the dark blue line), as this is a smaller bucket. We said before that we are at roughly 1999/2000 levels of total unemployment, but the 35-44 age bracket is at 1990 levels, and the 25-34 group is back to 1983 levels! The Boomers were born in 1946 through 1964 (according to the Census), so the waves you see in the chart show this generation rolling from one age bucket to the next. The only group above that is managing employment growth is the 55 and older bracket, and we’re willing to guess that it’s more structural than cyclical, as the boomers roll more heavily into this bucket. You’ll notice that the decline in the 35-44 age group began in the late 1990’s, and is clearly more structural than cyclical. The more worrisome part is that the largest age bucket, the 45-54 year olds, appear to have started their structural population decline. This will of course be offset by growth in other buckets, as well as a cyclical recovery in the economy and employment as a whole, but it points to a future problem of a stagnant (or possibly shrinking) pool of available labor.
This demographic shift makes the falling employment rate look no less devastating, but at least it explains the fact that we may be returning to a more “normalized” environment. It seems that we are slowly returning to an employment rate that will look more like the period before 1970, when we didn’t have a working population that included the pig in the python generation. This begs a whole new series of questions related to the future makeup, cost and productivity of the American work force, but that is a topic for another Salvo.
Posted in Current Events, Labor Markets | No Comments »
January 8, 2010
After a while, it’s tough to find new things to say on Nonfarm Payroll day. Coverage of this event is exhaustive, so we thought we’d try to add as few words as possible and update a few of the payroll graphs that we’ve looked at in recent months.
First, the good parts. Many strategists are looking at temp workers to be a leading indicator for total nonfarm, and the economy in general. And temp workers have been on the rise for the last 5 months.
As you can see, temp hiring led the way in the past two cycles, but we don’t have much of a historical record for this data series. You’ll notice that temporary workers corrected significantly more than total payrolls, and actually stands well below its lowest level of the previous downturn, whereas total nonfarm has yet to reach it.
Not sure if you want to call it a bright spot, but job leavers as a percentage of unemployed workers has ticked up slightly in the December release. An increasing number of workers willing to leave one job in search of another one would be an encouraging sign. However, it may be difficult to notice the small increase in the midst of all the historical context that we’ve provided in the graph below.
Now the not-so-good parts.
The average duration of unemployment is still spiking to new highs, and it now stands at 29.1 weeks.
The headline unemployment rate stayed at 10%, despite the -85,000 print on jobs for the month. However, the more broadly measured U6 unemployment rate (which includes marginally attached workers and workers employed part time for economic reasons) ticked up to 17.3% from 17.2% in November (but below October’s 17.4% reading).
Last but not least, as we wait for the V-shaped recovery to take hold, we watch the following chart:
If there is to be a vicious snap back, it is still out there on the horizon somewhere.
Posted in Labor Markets | 1 Comment »
January 5, 2010
Austrian business cycle theory is intensely focused on inflation as a byproduct of attempts by central banks to artificially stimulate the economy through the tools of monetary policy, i.e., lowering interest rates and increasing money supply. We have low rates, but money supply growth isn’t off the charts. Through December 21, 2009, the MZM monetary aggregate grew 4.7% year-over year and fell at an annualized rate of 1.9% over the last three months, and the narrower M2 grew 3.5% and 3.9% year-over-year and over the last three months, respectively. We do have, however, over a trillion dollars in excess reserves in our nation’s banks. Here’s a big assumption: Assume, for the moment (as many people are doing nowadays), that this is a proxy for the Austrian problem and therefore a precursor to inflation. Austrian economist Ludwig von Mises reminds us that inflation won’t be instantaneous or equal across all goods and services. He said, “Changes in the structure of prices brought about by changes in the supply of money available in the economic system never affect the prices of the various commodities and services to the same extent and at the same date.”
According to this perspective, if the government were to print money there would be a general yet uneven increase in prices of goods and services. It hasn’t happened yet, but it still might, particularly if those reserves make their way into the economy rather than stay stuffed in the banks’ mattresses on deposit at the Fed (where the Fed wants to keep it by paying interest on it, or converting it into term deposits, or drain it through the repurchase markets). We would suggest that von Mises’ dictum is instead taking place in financial assets. Consider the government’s balance sheet expansion that began in earnest in September 2008. As the graph below illustrates, this move preceded surges in the prices of gold (arguably a financial asset), then high yield bonds, then stocks.
 Source: Bloomberg We don’t want to confuse liquidity with monetary stimulus. In the latter condition, the Austrians will likely be right. In the former-which we have now-bubbles are created as liquidity seeks out higher returns by going out the risk curve.
Posted in Monetary Policy | 6 Comments »
|
|
