While Treasuries are said to have no default risk as the
Federal Reserve (Fed) can always print money to pay off the debt, hidden
risks might be lurking. As oxymoronic as it may sound, the biggest risk to
the economy and the U.S. dollar might be, well, economic growth! Let us
The U.S. government paid
an average interest rate of 2.046% on the $11.0 trillion of Treasuries
outstanding as of the end of November. Treasuries include Bills, Notes,
Bonds and Treasury Inflation-Protected Securities (TIPS). At 2.046%, the cost
of carrying the Treasury portfolio currently costs the government $225
billion per annum; about 6% of the federal budget was spent on servicing the
national debt. 1
While total government debt has ballooned in recent years,
the interest rate paid by the government on its debt has continued on its
We only need to go back to the average interest rate paid
in 2001, 6.19%, and the annual cost of servicing Treasuries would triple,
paying more than Greece as a percentage of the budget. Not only would other
government programs be crowded out, the debt service payments might likely be
considered unsustainable. Except for the fact that, unlike Greece, the Fed
can print its own money, diluting the value of the debt. In doing so, the
debt could be nominally paid, although we would expect inflation to be
substantially higher in such a scenario.
These numbers are no secret. Yet, absent of a gradual, yet
orderly decline of the U.S. dollar over the years – with the occasional
rally to make some investors believe the long-term decline of the U.S. dollar
may be over - the markets do not appear overly concerned. Reasons the market
aren’t particularly concerned include:
- The average interest rate continues to trend
downward. That’s because maturing high-coupon Treasury securities
are refinanced with new, lower yielding securities.
- Treasury Secretary Geithner has diligently
lengthened the average duration
of U.S. debt from about 4 years when he took office to currently over 5
For the U.S. government, a longer duration suggests less
vulnerability to a rise in interest rates, as it will take longer for a rise
in borrowing costs to filter through to the average debt outstanding. The
opposite is true for investors: the longer the average duration of a bond or
a bond portfolio one holds, the greater the interest risk, i.e. the risk that
the bonds fall in value as interest rates rise.
Operation Twist to hide interest risk
Debt management by the Treasury only tells part of the
story on interest risk. When the Treasury publishes “debt held by the
public,” it includes Treasuries purchased in the open market by the
Fed. By engaging in “Operation Twist”, the Federal Reserve
stepped onto Timothy Geithner’s turf, manipulating the average duration
of debt held by the private sector. Notably, the private sector holds fewer
longer-dated bonds, as the Fed has gobbled many of them up.
However, investors may still be exposed to substantial interest
risk in their overall fixed income holdings as, in the search of yield, many
have doubled down by seeking out longer dated and riskier securities.
The Fed, many are not aware of, employs amortized cost
accounting, rather than marking its holdings to market, thus hiding potential
losses should interest rates go up and its portfolio of Treasuries and
Mortgage-Backed Securities (MBS) fall in value.
Quantitative easing to increase interest risk
Whenever there’s a warning that all the money
created by the Federal Reserve is akin to printing money, some dismiss these
concerns as the money created out of thin air to buy securities has not
caused banks to lend, but park excess reserves back at the Federal Reserve.
As of December 14, 2012, $1.4
trillion in excess reserves is parked at the Fed. Substantial interest
risk might be baked into reserves:
Consider that the Fed has been paying about $80 billion in
profits to the Treasury in recent years. Think of it this way: the more money
the Fed “prints”, the more (Treasury & MBS) securities it
buys, the more interest it earns. That’s why Fed Chair Bernanke brags
that his policies have not cost taxpayers a cent, even if the activities may
put the purchasing power of the currency at risk. Now, Bernanke has also
claimed he could raise rates in 15 minutes. In our assessment, it’s
most unlikely he would do so by selling long-dated securities; instead, in an
effort to keep long-term rates low, the most likely scenario is that the Fed
will pay a higher interest rate on reserves. Up until the financial crisis
broke out, the Federal Reserve would have intervened in the Treasury market
by buying and selling securities to move short-term interest rates. In the
fall of 2008, the Fed was granted the authority by Congress to pay interest
As interest rates rise, not only will Treasury pay more
for debt it issues, it may also receive less from the Fed. Interest rates
would have to rise to about 6% for the entire $80 billion in
“profits” to be wiped out assuming a constant $1.5 trillion in
reserve balances ($1.4 trillion in excess reserves and $0.1 trillion in
required reserves that also receive interest); that assumes, the Fed does not
grow its balance sheet in the interim (in an effort to generate more
“profits” for the Fed) and would not reduce its payouts in the
interim as a precaution because bonds held on the Fed’s books may be trading
in the market at substantially lower levels.
Should interest rates move up, the Treasury may no longer
be able to rely on the Fed to finance the deficit (while the Fed denies the
purposes of its policies is to finance the deficit, the Fed is buying a
trillion dollars in debt as the government is running a trillion dollar
Biggest risk: economic growth?
In our surveys, inflation tends to be on top of
investors’ minds, no matter how often government surveys show us that
inflation is not the problem. Should inflation expectations continue to rise
– and a reasonable person may be excused for coming to that conclusion
given that the Fed appears to be increasingly focusing on employment rather
than inflation – bonds might be selling off, putting upward pressure on
the cost of borrowing for the government.
But if we assume inflation is indeed not an imminent
concern (keep in mind that the Fed is also buying TIPS and, thus, distorting
important inflation gauges in the market), we only need to look back at the
spring of this year when a couple of good economic indicators got some
investors to conclude that a recovery is finally under way. What happened?
The bond market sold off rather sharply! A key reason why the Fed is
increasingly moving towards employment targeting is to prevent a recurrence,
namely a market-driven tightening, pushing up mortgage costs.
The government should be grateful that we have this
“muddle-through” economy. Let some of that money that’s
been printed “stick;” let the economy kick into high gear. In
that scenario, the “good news” may well be reflected in a bond
market that turns into a bear market.
Historically, when interest rates move higher in an
economic recovery, the U.S. dollar is no beneficiary because foreigners tend
to hold lots of Treasuries: should the bond market turn into a bear market,
foreigners historically tend to wait for the end of the tightening cycle
before recommitting to U.S. Treasuries.
The point we are making is that for bonds to sell off and
the dollar to be under pressure, we don’t need inflation to show its
ugly head; we don’t need China or Japan to engage in financial warfare
by dumping their Treasury holdings. All we may need is economic growth! And
while Timothy Geithner has studiously been trying to extend the average
duration of U.S. debt, Ben Bernanke at the Fed has thrown him a curveball.
Perception is reality
One only needs to look at Spain to see that a long average
duration of government debt is no guarantor against a
debt crisis. Spain has an average maturity of government debt of 6 years, yet
it does not take a rocket scientist to figure out that borrowing at 6% in the
market is not sustainable given the total debt burden. As such, markets tend
to shiver when confidence is lost, even if, technically, governments could
cling on for a while when the cost of borrowing surges.
History may repeat itself
It was only 11 years ago that the US government paid and
average of 6% on its debt. Sure the average cost of borrowing has been coming
down. But no matter what scenarios we paint, if the average cost of borrowing
can come down to almost 2% from 6%, we believe it is entirely possible to
have the reverse take place over the next 11 years. Given the additional
risks the Fed’s actions have introduced, the timing could well be
condensed. But even if not, we believe it is irresponsible for policy makers
to pretend interest rates may stay low forever (except, maybe, Tim
Geithner’s steps to increase the average maturity of government debt;
but as pointed out, his efforts may be overwhelmed by those of the Fed).
What’s so sad about the discussion about the
so-called fiscal cliff is that even the initial Republican proposal results
in approximately $800 billion deficits each year. Financed at an average 2%,
this would add over $900 billion in interest expenses over ten years;
financed at an average 4%, it would add almost $2 trillion in interest
expense over ten years. Mind you, this is a politically unrealistic,
conservative proposal. Democrats pretend we don’t even have a long-term
sustainability problem, only that the wealthy don’t pay their fair
In our humble opinion, both Republicans and Democrats are
distracted. In many ways, the simultaneous increase in taxes and cut in
expenditures of the fiscal cliff is akin to European style austerity: if the
cliff were to take place in its entirety, we would a) suffer a significant
economic slow down; b) continue to run deficits exceeding 3% of GDP before
factoring in any slowdown; and c) still not have fixed entitlements.
While our discussion focused on $11 trillion in Treasury
securities, the so-called “unfunded liabilities” go much further
than the $5 trillion in accounting liabilities set aside. Depending on the
actuarial assumptions, unfunded liabilities may be as high as $50 trillion to
$200 trillion or higher.
In our assessment, the only way to tame the explosion of
government liabilities over the medium term is to tame entitlements. But it
is very difficult to cut back on promises made. As Europe has shown us, the
only language that policy makers understand may be that of the bond market.
As such, unless and until the bond market imposes entitlement reform, we are
rather pessimistic that our budget will be put on a sustainable footing. Put
another way, things are not bad enough for policy makers to make the tough
Different from Europe, however, the U.S. has a current
account deficit. As a result, a misbehaving bond
market may have far greater negative ramifications for the dollar than the
strains in the Eurozone bond markets had for the Euro. In the Eurozone, the
current account is roughly in balance; while there was a flight out of weaker
Eurozone countries, that flight was mostly intra-Eurozone towards Germany and
Northern European countries.
So while the default risk of U.S. Treasuries may be less
than that of Eurozone members, the risks to the purchasing power of the U.S.
dollar might be substantially higher. On that note, while the Fed has
indicated to buy another approximately $1 trillion in assets over the next
year, we would not be surprised to see the balance sheet of the more
demand-driven European Central Bank shrink as some banks pay back loans from
the Long-Term Refinancing Operation (LTRO) early.