February 5, 2013
The 2013 Economic Outlook – Bright Sunshine for the U.S., Periods of Cloud Abroad
Warning: Do not attempt reading the entirety of this commentary without the aid of your stimulant of choice. I apologize for the length of the commentary, but believed it necessary in order to inform new readers of my basic approach to macroeconomics. In future shorter (I promise) commentaries, there likely will be references to this one.
According to preliminary data, the U.S. economy’s growth in price-adjusted terms was 1.5% in 2012 on a Q4/Q4 basis. As things now stand, I believe that this pace will easily be exceeded in 2013. U.S domestic demand can be expected to accelerate. In addition, foreign demand for U.S. goods and services likely will be stronger. For financial markets, the implications of these expectations are that risk assets will outperform riskless ones and that the yield curve will steepen as the interest rates on longer-maturity fixed-income securities rise as the interest rates on shorter-maturity fixed-income securities remain anchored near their current levels by an unchanged Federal Reserve policy.
Let’s begin by discussing the principal driver of domestic demand – credit creation. It is important to understand that not all credit is created equal. Some credit is created figuratively out of “thin air.” An important implication of thin-air credit is that while its recipients (i.e., borrowers) will be able to increase their current spending, the grantors (i.e., lenders) of thin-air credit are under no compulsion to decrease their current spending. Hence, an increase in thin-air credit implies a net increase in nominal spending. The entities that are able to create credit out of thin air are central banks (i.e., the Federal Reserve in the U.S.) and the private depository institution system. Private depository institutions include commercial banks, savings banks and credit unions.
Before discussing the other broad category of credit, credit that is not created out of thin air, let’s describe how central banks and depository institution systems are able to create credit out of thin air. When a central bank purchases a security, it pays for that security by simply crediting the security seller’s bank account with an amount of “money” equal to the purchase price of the security. That increased amount of money the security seller now has was not transferred from some other entity’s bank account. Rather, it just “appeared” like manna from Heaven. The seller of the security to the central bank typically would replace the sold security with another security/loan, purchasing/funding the new security/loan with the manna from Heaven received from the central bank. Thus, indirectly, the central bank would have created new credit for the economy figuratively out of thin air. Even if not by law, by principle a central bank could grant a private entity a loan for the purchase of say, a house. This increase in credit would be created out of thin air.
In most financial systems, such as that of the U.S., depository institutions are required by law to hold only a fraction of the amount of the deposits on their books in the form of cash. Assume that depository institutions are required to hold at least 10% of their deposits as cash. (They would not be prohibited from holding an amount of cash greater than what is required by law if they so desired.) Let’s further assume that the depository institution system is made up of only one such institution, call it Jamie P. Morgan Bank (JPMB). If the central bank purchases $100 of securities from JPMB, recorded on the asset side of the balance sheet of JPMB and, thus, the depository institution system are a $100 decline in securities and a $100 increase in cash (technically, a $100 increase in JPMB’s deposit balances at the central bank). Under most circumstances, JPMB will want to replace the $100 of securities it sold to the central bank with another security or loan in an amount of $100. If it does so, there will be a net increase in credit to economy of $100. (The $100 of securities JPMB sold to the central bank represented credit that had previously been created. This credit was not extinguished when JPMB sold it to the central bank. Rather, the creditor changed from JPMB to the central bank.) Regardless of whether JPMB purchased a security or made a loan of $100, its deposits and, therefore, the deposits of the depository institution system, will have increased by $100. JPMB would then see its required cash “reserves” increase by $10, 10% of the $100 increase in deposits. Where would JPMB get the cash to meet its new higher required cash holdings? Well, because JPMB is the only depository institution in the system, the $100 it spent to purchase the security or make the loan comes right back to it in the form of the $100 deposit. Even though JPMB is required to hold an additional $10 in the form of cash, it still has $90 left that it could use to purchase another security or make another loan if it were profitable for it to do so. Assuming it is profitable to do, JPMB adds another $90 dollars of earning assets (loans and/or securities) to its balance sheet, which means that a total of $190 of net new credit has been created for the economy. Assuming it remained profitable and JPMB had the capital to support it, this process of credit creation could continue until a total of $1,000 ($100/0.1) of net new credit was added to the economy. So, by the central bank creating out of thin air $100 of “seed” money to the depository institution system, ultimately $1,000 of new credit potentially could have been created out of thin air for the economy. (It should be noted that if JPMB had some competitors, i.e., if there were some other depository institutions in the system, JPMB or its competitors would not be able to create new credit out of thin air individually because there is no guarantee that the funds created by an individual institution’s acquisition of a new earning asset would necessarily come back to that institution as a deposit. The funds could end up as a deposit at some other institution. But, the funds would remain in the depositary system as a whole and, therefore, the depositary system could potentially multiply $100 of seed money from the central bank into $1,000 of new credit for the economy created out of thin air.)
Now let’s distinguish between credit created out of thin air and credit created the old-fashioned way – through saving. Credit created through saving is not an impetus to the aggregate demand for goods/services/assets in the economy. Rather, credit created from saving transfers spending power from the grantor of this credit to the recipient. For example, households can increase their net provision of credit. They typically do this by curtailing their current spending and lending the funds that they otherwise would have spent to a borrower, perhaps a business or a government, that has a greater urgency to spend today than do households. Unlike central banks and depository institution systems, households and other nonfinancial entities are not generally capable of creating credit out of thin air. There is one scenario, however, whereby households and other nonfinancial entities could extend new credit that would result in a net increase in spending in the economy. This scenario would entail households funding their new lending, not by increasing their saving, i.e., not by curtailing their current spending, but by running down their deposits. If households are willing to hold fewer deposits than otherwise by substituting loan/securities /liabilities of financial intermediaries for these deposits, then the recipients of this credit extended directly from households or indirectly from households via financial intermediaries would increase their current spending and households need not, under these circumstances, curtail their current spending. Hence, in this special case, there would result a net increase in spending from the increase in credit even though this credit was not created by the central bank or the depository institution system.
So, let’s look at the historical relationship between changes in different categories of credit and changes in aggregate domestic spending on currently-produced goods and services in the U.S. economy. Plotted in Chart 1 are the year-over-year percentages changes in quarterly observations of combined Federal Reserve and depository institution credit (total “thin-air” credit) along with the year-over-year percentage changes in quarterly observations of nominal domestic purchases of currently-produced goods and services. Some of these goods and services were not produced in the U.S. but were imported. From Q1:1953 through Q4:2007, the correlation between percentage changes in total thin-air credit and percentage changes in nominal gross domestic purchases is 0.65 out of maximum possible 1.00. By “eyeballing” the chart, one can see that generally changes in total thin-air credit correspond closely with changes in nominal domestic purchases in the same direction from 1953 through 2007. This high positive correlation between changes in total thin-air credit and changes in nominal domestic purchases is what would be expected from the prior explanation. An increase in thin-air credit allows the recipients to increase their current spending while not requiring the grantors of this credit to curtail their current spending. Hence, an increase in thin-air credit carries with it a presumption that there will be an increase in domestic spending.
This positive relationship, however, went awry in 2008. In that year, total thin-air credit growth skyrocketed as the Federal Reserve’s balance sheet expanded in reaction to the financial crisis and nominal domestic purchases contracted by the most since the post-WWII era. The principal reason the positive relationship between changes in thin-air credit and changes in nominal domestic spending turned negative in 2008 is that a large part of the expansion in the Federal Reserve’s balance sheet was loans to foreign central banks, who, in turn, extended U.S. dollar-denominated funding credit to their constituent banks who were facing a U.S. dollar liquidity squeeze. Thin-air credit extended to a foreign central bank would not be expected to result in an increase in U.S domestic purchases. Subsequent to the unusual events in 2008, the positive relationship between changes in total thin-air credit and changes in nominal domestic purchases has re-established itself.
Now, let’s look at the historical relationship between percentage changes in credit extended by nonfinancial entities (e.g., households, nonfinancial businesses, governments, the rest of the world) and percentage changes in nominal gross domestic purchases, as shown in Chart 2. From Q1:1953 through Q4:2007, the correlation between these two series is 0.39 vs. 0.65 for total thin-air credit changes. Given that the bulk of credit extended by nonfinancial entities is funded by these entities curtailing their current spending, i.e., increasing their saving, this category of credit extension tends just to redistribute a given amount of spending rather than resulting in a net increase in spending. Thus, it would not be expected that there would be a close positive relationship between changes in credit extended by nonfinancial entities and changes in nominal domestic purchases.
Lastly, let’s look at the historical relationship between percentage changes in credit extended by financial intermediaries other than depository institutions (e.g., insurance companies, pension funds, hedge funds) and percentage changes in nominal gross domestic purchases, as shown in Chart 3. From Q1:1953 through Q4:2007, the correlation between these two series is 0.28 vs. 0.65 for total thin-air credit changes. Again, it would not be expected that there would be a close positive relationship between changes in credit issued by non-depository financial intermediaries and changes in nominal domestic purchases unless the ultimate providers of this credit, nonfinancial entities, were willing to substitute the liabilities of these non-depository financial intermediaries for their holdings of deposits, i.e., the liabilities of depository institutions.
So, let’s summarize what has been discussed so far. There are two broad categories of credit – credit that is created figuratively out of thin air and credit that transfers spending power from the grantor to the recipient. An increase in credit created out of thin air would be expected to result in a net increase in nominal spending on goods/service/assets. An increase in credit transferred from one entity to another would not be expected to result in a net increase in nominal spending on goods/service/assets, but rather just to redistribute a given amount of nominal spending.
Now that we understand that the behavior of thin-air credit plays a key role in the behavior of domestic spending in an economy, let’s discuss the recent and near-term expected behavior of thin-air credit in the U.S. This is shown in Chart 4. In the three months ended December 2012, there was an acceleration in the growth of combined Federal Reserve and commercial bank thin-air credit to an annualized rate of 3.4%. Both components of this thin-air credit aggregate contributed to this accelerated growth. In the latter part of 2012, the Federal Reserve announced that it would begin increasing its securities holdings by a net $85 billion per month for an undetermined time. Assuming other asset items on the Federal Reserve’s balance sheet were to remain relatively constant, this would imply an annual increase in the thin-air credit created by the Fed of $1.02 trillion. If this pace were maintained in 2013, it would represent an annual rate of increase in Federal Reserve created thin-air credit in excess of 35%. If commercial bank credit were to remain at its December 2012 level throughout 2013 and if the Federal Reserve’s balance sheet were to increase a net $1.02 trillion in the 12 months ended December 2013, then the 2013 December-over-December change in the sum of Federal Reserve and commercial bank credit would be 8.0%. To put this 8.0% year-over-year change in perspective, the median year-over-year change in this credit aggregate from December 1991 through December 2012 was 6.77%. So, what the Federal Reserve is doing with regard to increasing the size of its balance sheet – creating credit out of thin air – is likely to stimulate U.S. domestic spending quite significantly.
From examining Chart 4, we can see that this would not the first time in recent years in which the Federal Reserve’s balance sheet grew rapidly. For example, there was a spike in the3-month growth of the Federal Reserve’s balance sheet in the first half of 2011, only to see this growth dissipate in the second half of 2011. What might be different this time? The Federal Reserve recently announced more specific observable goals for its policy actions. The Federal Reserve has indicated that it does not intend to raise its policy interest rates until the unemployment rate, which stood at 7.9% in January, falls to 6.5% or unless the consumer inflation rate, which was 1.3% year-over-year in December, is projected to rise above 2.5% in the next one to two years. The Federal Reserve did not announce the same conditions for terminating its current pace of balance-sheet expansion, but one could infer that this round of Federal Reserve balance-sheet expansion is more economic-goal dependent than time/amount dependent, as was the case in recent years. So, I expect that the Federal Reserve’s current pace of net securities acquisitions of $85 billion per month will persist throughout most of 2013, if not all.
The Federal Reserve currently accounts for only about 20% of total thin-air credit. By far, the biggest component of total thin-air credit is commercial banking system credit. As shown in Chart 4, bank credit resumed steady, albeit historically weak, growth around mid 2011. In the three months ended December 2012, bank credit growth re-accelerated to 3.75% annualized. I expect that bank credit growth will accelerate further throughout 2013 as banks become more willing to extend new credit. But even if commercial bank credit were to grow at an annualized rate of 3.75% for all of 2013, along with the stepped up growth in Federal Reserve credit, as discussed above, growth in the sum of Federal Reserve and commercial bank credit is likely to be quite strong in 2013, which implies strong growth in domestic purchases of goods/service/assets.
There is a common view that bank credit growth has remained historically weak in recent years because of lack of demand for credit. I reject this view. Rather, I believe that banks have been reluctant to accommodate the demand for credit because of current or expected capital constraints. Banks need adequate capital to expand the amount of assets on their balance sheets. In 2009, after the worst financial crisis in the U.S. since the early 1930s, the U.S banking system experienced a crippling “evaporation” of capital. After that, regulatory authorities indicated that required capital ratios for banks would be rising by a to-be-determined amount. Even if banks currently had adequate capital to resume lending, some were uncertain about their future capital adequacy due to continued declines in real estate prices, both residential as well as commercial, and by the uncertainty as to what regulatory capital ratios would be. The recovery in U.S. real estate markets and the stabilization of real estate prices, if not increases in these prices, have provided more certainty to banks about their future real estate write-downs. And, of course, any lender would prefer to extend credit backed by collateral that is rising in price than collateral that is declining in price. In addition, the regulatory environment for banks has become more certain. This, plus a massive capital-raising campaign by banks in recent years, puts banks in a position to step up their lending.
Has there been unrequited demand for credit from the household sector? I argue that there is reason to believe so. Chart 5 shows the household debt-service burden – required principal and interest payments on outstanding household debt as a percent of after-tax personal income. With the general decline in interest rates, the household debt-service burden has plunged to its lowest level since the early 1990s. So, in terms of monthly required principal and interest payments, households are able to take on more debt.
Would any household want to take out a home mortgage today? If not now, when? Chart 6 shows the historical “yield” on owner-occupied housing vs. the cost of financing a house, the mortgage rate. The yield on owner-occupied housing is obtained by dividing the imputed rent on owner-occupied housing by the market value of owner-occupied housing (and multiplying by 100 in order to put it into percentage terms). In Q3:2012, the yield on owner-occupied housing stood at 7.26% vs. an effective mortgage rate of 3.72%. So, in Q3:2012, assuming you could qualify for a mortgage, you could have acquired an asset with a current yield of 7.26% and financed that asset at a borrowing rate of 3.72%. Sounds like a good deal, right? This differential between the yield on housing and the mortgage rate in Q3:2012 was 3.54%, the widest positive differential in the history of the series. In fact, this differential turned positive in Q4:2008 and has been trending higher to its current record high. So, why was there not the beginning of a sustained recovery in home sales until 2011? Because, although housing has been an attractive purchase in the past three years, banks have been reluctant to extend new housing financing until mid 2011. As Chart 4 shows, combined Federal Reserve and bank credit resumed growth in early 2011, about the same time that home sales began trending higher. I do not think that it is any coincidence that another credit-sensitive sector of household spending has experienced a recovery since thin-air credit started growing again – motor vehicle sales.
Now, there could be some intermediate periods of “cloudiness” in the U.S. economy during 2013 due to federal budgetary issues. The federal budget deficit declined in 2012, as shown in Chart 7, and it will decline further in 2013. Of course, the budget deficit is a function of government outlays as well as receipts. Also shown in Chart 7 is the year-over-year percentage change in the 12-month moving total of federal outlays. For example, the last data point in Chart 7 is the percent change in the sum of total federal outlays in the 12 months ended December 2012 vs. the sum of total federal outlays for the 12 months ended December 2011. As one can see from examining Chart 7, growth in federal outlays soared in 2009 at the depth of the recession. Since the end of the recession, however, growth in federal outlays has been considerably more subdued. In fact, there have been a number of 12-month spans when federal outlays have contracted. There are two reasons for the more subdued growth in federal spending in the past three years. Firstly, with the recovery in the economy, albeit a weak recovery, growth in some income-maintenance programs, such as unemployment insurance benefit payments, has slowed. Secondly, Congress passed legislation in the summer of 2011 that called for a reduction in federal spending vs. plan of more than $1 trillion over the next 10 years. This congressional governor on spending played a key role in slowing federal –outlay growth in calendar 2012. Federal revenue growth rebounded with the recovery in the economy after the last recession. Both household and corporate tax receipts were boosted by the return of GDP growth.
Growth in federal spending is set to come under further restraint in 2013. The sequestration of $1.2 trillion of federal outlays over the next 10 years is scheduled to kick in this March. Even if the sequestration legislation is modified, there is a likelihood of a further significant restraint on federal spending. Revenues will be increasing due to the end of the payroll-tax “holiday,” the higher marginal tax rates on upper-income households and the increased taxes associated with the Affordable [health] Care Act.
Will this combination of additional federal spending restraint and higher tax revenues have a material negative effect on growth in total domestic spending over the next year or so? Both theory and empirical evidence suggest it will not. The spending restraint and the higher tax revenues will result in a small federal deficit. By definition, this means that the federal government’s borrowing requirement will be less than otherwise. In turn, this implies that entities that otherwise would have been lending to the federal government will now find themselves with “excess” funds. There are three things these entities can do with these funds and two of them will offset spending reductions emanating from the restraint in federal-outlay growth or private-spending reductions emanating from the payment of higher taxes. One thing the entities with excess funds can do is lend these funds to households, businesses or other forms of government, who will then spend the funds. Alternatively, the entities with excess funds can spend the funds themselves. Lastly, the entities with excess funds can decide to simply hold them in the form of higher deposits. Only in this last case, where the demand for deposits rises, will there be no other spending to offset the reductions in spending emanating from the “tighter” fiscal policy.
That’s the theory as to why changes in fiscal policy have no material effect on total spending in the economy. Here’s the empirical evidence in Chart 8. The data in Chart 8 are the year-to-year percentage point changes in the budget stance of the federal government and year-to-year percent changes in nominal GDP. The changes in GDP are straightforward. But some explanation is in order for the changes in fiscal stance. The budgetary position of the federal government is affected by the behavior of the economy. When the economy is growing rapidly, tax revenues are boosted for a given tax-rate structure because household incomes and corporate profits are growing faster. Federal outlays tend to be more restrained when the economy is growing rapidly because income-maintenance expenditures, such as unemployment insurance benefit payments and food stamp expenditures grow more slowly. So, to test for the effect of changes in the budget situation on changes in GDP, the budget surplus/deficit needs to be adjusted for the cyclical effects on it just described. The Congressional Budget Office (CBO) publishes a series that adjusts the federal budget surplus/deficit for these cyclical effects. The CBO also calculates this cyclically-adjusted budget surplus/deficit as a percent of potential GDP in order to scale the budget concept. The red bars in Chart 8 represent the year-to-year percentage-point change in the cyclically-adjusted budget surplus/deficit as a percent of GDP. According to mainstream-media economics, a change in fiscal policy that would result in a larger budget surplus/smaller budget deficit (the red bars in Chart 8 become more positive or less negative) should be associated with slower growth in nominal GDP and vice versa. That is, according to mainstream-media economics, there should be a negative relationship or correlation between changes in the fiscal policy variable in Chart 8 and changes in GDP. If changes in fiscal policy are compared with changes in GDP contemporaneously (not shown in Chart 8), the correlation turns out to be positive. That is, when looking at changes in the fiscal policy variable and changes in GDP in the same year, an increase in the budget surplus/decrease in the deficit tends to be associated with an increase in GDP growth. Well, to be fair to the mainstream-media economics, it is reasonable to expect that there would be some lag between a change in fiscal policy and its effect on GDP growth. In order to get the negative correlation between changes in fiscal policy and changes in GDP predicted by mainstream-media economics, changes in fiscal policy have to lead changes in GDP by two years (the “-2” in the “red” title in Chart 8). An absolute value of .02 for a correlation is close enough to zero “for government work.” In other words, the data suggest that there is no meaningful relationship between changes in fiscal policy and changes in GDP. To re-iterate, both on theoretical and empirical grounds, I do not believe that the likely ‘tightening” in federal fiscal policy in 2013 will have a materially negative effect on growth in U.S. total domestic demand on an annual basis.
What’s the economic outlook for the rest of the world? Let’s start with the Chinese economic outlook. I forecast “sunshine,” but perhaps not as relatively bright as that for the U.S. The Chinese economy overheated in 2009-2010 as result of a surge in Chinese depository-institution credit, i.e., thin-air credit (see Chart 9). As a result of the overheating, Chinese consumer inflation accelerated. In response to these developments, the Chinese monetary authority, the People’s Bank of China (PBOC), took actions in 2010 and 2011 that pushed up short-term Chinese interest rates and slowed the growth in Chinese thin-air credit. As consumer inflation abated in 2012, the PBOC reversed policy, which resulted in a re-acceleration in Chinese thin-air credit growth. Although the PBOC is unlikely to countenance growth in Chinese thin-air credit as rapid as what occurred in 2009, I do expect that Chinese thin-air credit growth will be maintained at its current rate or a bit higher. This will result in a further re-acceleration in Chinese domestic-demand growth. A little-appreciated fact, but China is a huge importer of raw materials and semi-finished goods, such as electronic components. This implies that the expected acceleration in Chinese domestic-demand growth will stimulate growth in other economies via their export sectors.
One economy that will benefit from increased exports to China will be the Japanese economy. Japanese domestic demand is likely to receive a boost from Japanese monetary policy. The new Japanese prime minister, Shinzo Abe, will name the new governor and two deputy governors of the Japanese central bank, the Bank of Japan (BOJ), in March. The new governor of the central bank is expected to implement a more aggressive accommodative monetary policy. Given that Japanese short-term interest rates are near zero, this more aggressive accommodative monetary policy would be expected to manifest itself by a faster expansion in the BOJ’s balance sheet. So, Japanese thin-air credit is set to grow faster in 2013. Regardless of the growth in thin-air credit, the Japanese economy’s potential real growth rate is not what it was 30 years ago because Japanese labor force is declining because of the rapid aging of the population. Any economy’s potential growth rate, be it Japanese or American, is dictated by the growth in its labor force and the growth in the productivity of its labor force. Unless Japanese labor productivity growth were to experience some extraordinary permanent boost, the contracting Japanese labor force will limit aggregate Japanese real economic growth. Of course, real Japanese per capita income could still grow as rapidly as it did 30 years ago.
If there is a persistent cloud in the 2013 economic outlook, it hangs over Europe. The European banks have endured two major damaging “storms” since 2008. The first storm originated in the U.S. with the bursting of the American housing bubble and the failure of Lehman Brothers. The second storm to hit the European banks was spawned at home – the fiscal calamities of the southern European governments. Both storms reduced the capital of European banks, inhibiting their ability to create normal amounts of thin-air credit. Although the European Central Bank (ECB) has taken bold steps to supply liquidity to European banks in order to prevent their insolvency, the ECB has been unable due to political reasons and/ or unwilling due to a lack of understanding to expand its balance sheet so as to maintain a normal rate of growth in eurozone thin-air credit (see Chart 10). Unless the ECB decides to step up its provision of thin-air credit or eurozone depository institutions miraculously find the capital to expand their credit creation, the European economy as a whole will remain mired in a mild recession in 2013.
So, as CNBC’s Maria Bartiromo might ask, “How do I make money with all this?” Well, the expected acceleration in the growth of thin-air credit globally except for Europe will accelerate global spending on goods, service and assets. The acceleration in the growth of thin-air credit also will put a “whiff” of higher inflation in the global atmosphere. The implication of stronger global aggregate demand with the abatement of deflationary pressures is bullish for risk assets such as equities, “junk” bonds, real estate and commodities. It is bearish for assets with less credit risk such as investment grade corporate and government bonds. I believe that investment grade bonds in the U.S. are particularly at risk. If U.S. thin-air credit grows by 8% or more in 2013, which is quite likely given the Federal Reserve’s current rate of net securities acquisitions, then the U.S. unemployment rate is likely to fall faster and U.S. inflationary pressures are likely to build faster than the Federal Reserve currently is forecasting. This will bring forward in time expectations as to when the Federal Reserve will begin pushing up short-term interest rates. Thus, the U.S. government securities yield curve will steepen in 2013 as bond yields rise with money-market interest rates still anchored by Federal Reserve policy. I believe that the Federal Reserve will start pushing up short-term interest rates in 2014, not 2015 as is generally expected, and, after a few “baby-step” interest rate increases, the Federal Reserve will then make some “giant-step” interest rate hikes. In other words, I believe the behavior of bond yields in 2014 will “rhyme” with, if not repeat, their 1994 behavior.
Paul L. Kasriel