Market participants increasingly expect the Federal Reserve to begin unwinding its balance sheet during the second half of 2017, and many have speculated that the agency mortgage-backed securities (MBS) market – and the U.S. housing market more broadly – could suffer as a result.
For context, the Fed owns approximately $ 1.75 trillion of agency MBS, accounting for roughly 29% of the market and an even larger percentage of tradable float (agency MBS actively circulated in the market and not held by banks, government-sponsored enterprises, etc.). While those numbers are significant, we believe the agency mortgage market is unlikely to exhibit massive volatility as the Fed begins to reduce its unprecedented footprint, for four key reasons:
A mindful Fed. The Fed is actively trying to avoid disrupting financial markets as it reduces accommodation, as evidenced by its carefully telegraphed, measured hiking of short-term interest rates. We expect the Fed to take a similar approach to balance sheet reduction, telegraphing its moves and exiting extremely gradually, likely by tapering reinvestments as opposed to ending reinvestments entirely or selling bonds outright.
Reasonable starting valuations. Agency MBS have been cheapening significantly for nearly two years and are now close to historically fair value. We believe 2010 – the last year in which the Fed was not adding to or reinvesting its MBS holdings – provides a useful benchmark for analyzing current MBS valuations. In the 2010 environment, agency MBS were approximately 20 to 35 basis points (bps) cheaper than they are currently (on an option-adjusted spread basis). So, is a short position smart today? Likely not, as agency MBS provide roughly 35 bps of additional carry per year relative to “risk-free” interest rates. In our view, unless they can size and time short positions perfectly (tough to do), investors are unlikely to make money shorting agency MBS at current levels.
Attractive relative valuations. Not only have MBS cheapened over the past two years, they’ve done so while most other spread sectors were richening significantly. For example, U.S. investment grade credit spreads are roughly 60 bps tighter today relative to 2010 (as measured by the Bloomberg Barclays U.S. Credit Index). Moreover, we’re now seven years further into an expansionary cycle, and agency MBS tend to outperform credit in the latter half of expansions (and recessions).
Favorable technicals. We see several key technical factors likely to benefit MBS as the Fed reduces its balance sheet:
- Net agency MBS issuance totaled roughly $ 300 billion last year and is on pace to reach that figure again in 2017 (at roughly $ 220 billion to date). Importantly, about $ 110 billion of issuance last year consisted of cash-out refinancing, and we would not expect that pace to continue if interest rates rise from current levels.
- The government-sponsored entities (GSEs) – Fannie Mae and Freddie Mac – have sold roughly $ 70 billion of agency MBS per year since 2009 but are nearing the end of their MBS asset sales. While Fed sales are likely to exceed the previous pace of GSE sales, the markets have been dealing with a “taper” of sorts for years, and the Fed will now be the one selling as the GSE portfolios are no longer.
- Demand from overseas buyers is likely to stabilize or pick up from current levels, given the robust demand for high quality spread assets. Foreign holdings of agencies, for example, peaked in 2008 at $ 890 billion but have since fallen to $ 255 billion, while also shrinking as a percentage of reserves. We expect this trend to slow and potentially reverse as the Fed provides more clarity on the unwinding of its balance sheet.
- Bank demand may also rise. Excess reserves in the banking system will likely decline as the Fed reduces its balance sheet, which will require banks to purchase more high quality liquid assets (HQLA) over time. We expect agency MBS to serve a key role for banks in this regard.
The bottom line
We view MBS as relatively fairly valued at current levels, and the U.S. government conservatorship, liquidity and carry of agency MBS may provide attractive diversification and other benefits for credit-heavy portfolios in the current environment. As the Fed’s balance sheet unwind nears, it’s important to take all factors into account when assessing the risk/reward profile of the MBS sector. Given the four factors above – a mindful Fed, reasonable starting valuations, attractive relative valuations and favorable technicals – we think those expecting a “bubble to burst” as the Fed balance sheet shrinks may be greatly underwhelmed.
For more of our insights on Fed policy and its impact on the mortgage markets, see our latest Secular Outlook, “Pivot Points.”
Daniel Hyman is co-head of the agency mortgage portfolio management team and lead portfolio manager of the PIMCO Mortgage Opportunities Strategy.