While mom and pop investors and BTFDers (if not so much hedge and mutual funds and other "smart money") have been delighted by the latest V-shaped surge in stocks, it has come as we have repeatedly shown...
... at the expense of collapsing long-term yields as another central bank liquidity tsunami is priced in. In fact, early Friday both 10Y and 30Y US Treasury yields plunged to new all time lows, a signal which at any other time would suggest a deflationary tsunami is about to be unleashed, but in this case simply meant that another bout of central bank generosity was coming to prop up risky assets in the aftermath of Brexit.
The problem is that while stocks can - for now - ignore this historic divergence, which has pushed the S&P back to just shy of all time highs while bond yields are at all time lows, one major market participant can no longer pretend to not notice what is going on. We are talking about pension funds, who according to Bank of America are about to "throw in the towel" and capitulate on the de-risking of their portfolios, unleashing the next major buying spree on the long end, in the process likely pushing the 10Y to 1% or even much lower.
As BofA's Shyam Rajan writes, bull flattening of yield curves is rarely good news to anyone – but defined benefit pension plans are most leveraged to this pain. According to the most recent Milliman estimate, the average funded ratio of the top 100 US corporate defined benefit pension plans already had dropped to 77% by end of April. Since then, 30y rates dropped another 50bp and corporate spreads have tightened. While the asset side has provided some relief given that equities are hanging on to the highs, we think it is safe to assume that funding ratios over the last month have now reached the lows seen in 2012 – a rather sobering thought given that the equity rally of 70% since then has meant nothing and has been subsumed by the rate decline. The dominant factor of pension funding gaps has been the move in rates, as Chart 2 makes clear.
According to BofA there are five reasons why capitulation is more likely now.
Talking about pension capitulation seems counterintuitive when funded ratios are at record lows on the heels of a significant decline in rates. After all, why would a pension manager hoping for mean reversion at the beginning of the year feel forced to throw in the towel at these levels? 10y rates have been here before, funding ratios have been this low before, and this is not the first time for a flight to quality out of Europe and Japan into the US. What makes this time different? We identify five reasons (three macro, two pension-specific) that make capitulation this time around much more likely:
Here are the reasons:
1. Longer term growth
The key difference from a few years ago is the formalization of the “new normal” in the markets. Global estimates of neutral real rates are much lower, the Fed’s estimate of the long-run rate has dropped nearly 100bp, and the yield curve in itself sends a bleak message (Chart 2). While in 2013, 10y rates reached similar levels and the market pushed out the Fed nearly three years, it remained optimistic for the long run. Terminal rates were priced to be north of 3.5%, with forward inflation expectations above the Fed's target. Today, every intermediate forward beyond 3y1y is 50-200bp lower than the lowest point in 2013. There is greater understanding that the yield moves are not temporary but a glaring reflection of the new normal across the globe
2. Inflation expectations
The inflation market signals the same pessimism. Chart 4 shows inflation expectations across global markets relative to central bank inflation targets. Zero on the chart indicates that the markets on average expect the central bank to hit its inflation target over 30 years, while negative indicates a miss to the target. Among the markets that priced-in positive inflation risk premia in late 2013, almost all now project their central banks to miss their targets over 30 years.
3. The gravitational pull of negative yields
Any assumed lower bound for rates has been thrown out the window given the moves to negative rates. Nearly 33% of the BofAML Developed Market (DM) sovereign index now trades negative in yield, and the US makes up nearly half of the positive yielding assets available to investors. The long end of the US curve remains cheap to European and Japanese investors (on unhedged or partially hedged basis) who are getting pushed out of their domestic sovereign markets because of QE. Fundamentally, the macro rationale for the ECB and BoJ to stop QE is years away, and flow-wise, the safe haven scarcity problem motivating flows into the US is here to stay.
4. Variable PBGC premiums
The penalties for underfunding have increased markedly since 2012 and are scheduled to increase even more. Based on the latest budget act, for each $1,000 of underfunding, variable rate Pension benefit Guarantee Corporation (PBGC) premiums increase from $30 to $33 in 2017, $37 in 2018, and $41 in 2019. Nearly 33% of the BofAML Developed Market (DM) sovereign index now trades negative in yield Essentially, the top 100 corporate defined benefit pension plans will be paying at least ~$20bn/year in variable premiums by 2019 if current levels of underfunding remain.
5. Borrowing to solve the pension tension
The greatest impediment to pension risk transfers (used loosely as a term for offloading some or all of the pension risk to an insurance company who then fund them with annuities) is the need to raise cash to bring funding ratios to par before passing it on to the insurance company. In this regard, low corporate bond yields (and ECB corporate buying) actually could help. Consider the BofAML Corporate Master index – the effective yield of single A rated corporates is 2.52% and that of BBBs is 3.4%. The relative tradeoff now between issuing debt vs paying 3-4% of variable premiums is increasingly attractive – corporates could consider issuing debt to make pensions whole and probably come out with a net positive after tax savings on debt interest (Chart 6). While issuing debt to buyback stock has been a popular strategy, issuing debt to make pensions whole could be the next trend in fixing balance sheet risk. A broader discussion of this trade-off is beyond the scope of our piece, but it reinforces the point that despite wider funding gaps, there is likely a greater incentive or larger fear that is likely to motivate pension de-risking in the coming months.
To be sure, the current collapse in yields has precedented: in 2011, rates declined more than 100bp, curves flattened by more than 75bp, the large scale downgrade of European bonds left Treasuries as the only choice, and there was greater acceptance that 10y yields probably wouldn’t go back to 4% anytime soon. This move prompted significant pension de-risking in 2012 – evident in the two large risk transfer trades (GM and Verizon), widening long-end swap spreads, and significant increase in Treasury holdings of defined benefit pension plans (Chart 7).
As Bank of America summarizes, a capitulation at this point seems inevitable:
Today, all of the above is amplified. Treasuries make up nearly 50% of the positive-yielding DM sovereign bonds; curves are 100bp flatter; and there is a greater likelihood that 10y yields probably won’t go back to even 2.5%. We would expect a bigger capitulation by pension managers in the coming months/years.
How big of a market are we talking about here in case there indeed is a capitulation be? Well, in a word - massive. It’s a $3 trillion trade.
For the rates market, the significance of this acceptance phase by pensions cannot be understated, in our opinion. A $3 trillion industry running a $500 billion funding gap and a significant duration gap waking up to reality is likely to have major implications for the market. The nature of the de-risking is less important but could amplify the impact. In the simplest de-risking scenario, pension managers would stop underestimating the perceived lower bound for US rates and be more aggressive in using rate sell-offs to close duration gaps. In the extreme case, entire pensions could be offloaded from corporate balance sheets to insurance companies (increasingly like the UK, Exhibit 1)–generating significant demand for long-end duration during such transactions. One only needs to look at the long end of the UK rates market to see the significance of pension demand (Chart 9). Note that the UK regulation on inflation protection for pensions is more stringent, leaving the impact primarily on real yields. A similar move in the US is likely to be more evenly divided between reals and breakevens.
The final question: what will be the market impact of the capitulation:
Flatter curves, positive sign for swap spreads & long-end balance sheet trades. Ultimately, the de-risking of pensions whether via a full risk transfer or not would have significant implications for the rates market. Defined benefit pension holdings of USTs still stands at a rather low 6% of total assets (Chart 8)
- Rates: It would add to the long list of buyers (Japanese, European investors, index shorts) eager to add duration on any modest sell-off in US rates. This would limit any sell-off in rates to short-lived, positioning-led squeezes higher in yields, following which a flood of demand would take over. Active hedging by pensions also
is likely to keep receiver skew in longer tails rich.
- Curve: Any sizable de-risking is likely to be a flattener. Even if corporations issue to shore up pensions, which then subsequently de-risk, the net impact would likely be a flattener, in our view. This is because issuance would be skewed toward the belly of the curve where demand dominates, while the de-risking flow happens in the long end of the curve. While this is a long-term theme, it should help our tactical flattening call on the curve.
- Spreads and strips: Long-end demand from pensions also would be the welcome sign of relief the long end of the spread and coupon-strip curve need. The lack of consistent real money demand combined with $150bn in 30y UST supply every year and lack of dealer appetite to police these relationships has in short been the primary driver of tightening of long-end spreads and cheapening of coupon strips. Some 46bp of extra yield in USTs over swaps and 25bp of extra yield in c-strips over p-strips should look extremely attractive to investors settling for below-2% yields.
What all of this means in simple, numeric terms for the two securities everyone is most familiar with, namely the 10Y and the 30%? It means look for the 10Y Treasury to drop under 1% while the 30Y plunges to 1.50% or lower, as the entire world slowly but surely turn Japanese, where incidentally, the world's largest pension fund - the GPIF - just lost $43 billion in the past quarter as a result of the failure of Abenomics and its hail mary attempt to offset billions in underfunding using a monetary policy gimmick. Similar losses are coming to pension funds much closer to you next.