Absolute return
21 Oct 2016

What a trillion dollar outflow means for libor rates

We have reached an important milestone for the mechanics of short-term US dollar interest rates, with the passing of the deadline for prime money market fund (MMF) reforms on 14 October.

  102 viewed

We have reached an important milestone for the mechanics of short-term US dollar interest rates, with the passing of the deadline for prime money market fund (MMF) reforms on 14 October.

As we discussed in a note published in August, the reforms require such funds – which invest primarily in corporate debt securities – to switch to floating net asset values (NAVs); hold more very short-dated paper; and, on occasion, impose redemption gates and liquidity levies.

This deadline has had material implications on the functioning of US dollar money markets, given that prime MMFs have historically been large buyers of bank commercial paper (CP) – the yield on which is an important input into the setting price of Libor. (Libor, in turn, serves as the floating index for a large proportion of interest-rate derivative and ‘real economy’ instruments, such as mortgages).

The impact of the reforms on assets under management in the prime MMF universe has been dramatic. Since this time last year, the combined outflow from institutional and retail prime MMFs has been over US$1 trillion (from US$1.419 trillion to US$385bn now, according to iMoneyNet and JPMorgan), with US$61bn leaving in the second week of October alone. This has led to a sharp reduction in the average maturity of such funds’ holdings, with the weighted-average maturity of securities held by institutional prime MMFs now at just 11 days, down from 31 days a year ago.

In other words, over two thirds of the industry’s assets under management has been withdrawn, and those remaining assets appear to be managed very ‘close to home’ in the event of further outflows in the weeks ahead.  

In light of this, it is perhaps surprising that short-end money market rates have not reacted more strongly, and indeed the process appears to have been relatively smooth with regard to Libor rates and other funding market indicators. Aside from a fairly messy quarter-end period in the last week of September (quarter- and year-end periods have recently been a source of periodic instability, given the impact of regulation) there is little evidence of funding market pressure.


Libor rates have not risen to any meaningful degree since the initial move higher over the summer after one strips out the impact of increased investor expectations for an interest-rate hike by the US Federal Reserve later this year. Indeed, since our previous note, three-month Libor has gone from trading 40bps above three-month overnight index swap (OIS) rates (a purer gauge of market expectations for central bank policy rates) to just 41bps, after the deadline passed.

Source: Bloomberg, as at 18 October

And while one-year interest-rate swap rates, based on three-month Libor, have increased from 0.90% to about 0.97% since our previous update, one-year OIS rates have increased by a greater amount: from 0.49% to about 0.59%. This implies the move is wholly down to increased expectations of policy rate increases, rather than a surge in Libor rates due to funding-market pressure.

So what explains this relatively benign reaction in Libor rates to what appears to be a very significant reduction in demand for bank funding? Several factors have likely contributed:

  1. The reform date was well-telegraphed, and banks that were very reliant on US dollar CP funding (such as Japanese lenders) have had ample time to arrange other sources of funding. (In the case of Japanese banks, however, there is still very clearly a US dollar asset-liability mismatch, which should require ongoing management). This means that while some banks had opted to ‘pay up’ for term funding in CP markets over the summer, they have not been forced to do so as their maturing CP securities have rolled off (and not been rolled) into the reform deadline.
  2. Some prime funds adopted floating NAVs ahead of the deadline in order to give themselves and their investors ample time to deal with the changes, which has smoothed the process out over some weeks instead of being a sudden event.
  3. There is some evidence to suggest that CP yields are only a relatively minor input into the three-month Libor rate-setting process, perhaps contributing about 25% of the information that bank treasury desks call upon, with the remaining 75% a function of other factors such as repo rates (the cost of borrowing against Treasury securities over the short-term), OIS rates, and interpolation from longer-dated issuance (according to Intercontinental Exchange, or ICE, the LIBOR administrator). In the absence of wider stress, therefore, banks have not passed on higher CP yields into their Libor submissions.

But does this mean we are out of the woods – and will three-month Libor return to its 2013-2015 average of setting 10-15bps over three-month OIS? The market is pricing some degree of normalisation, but not a full reversal of the recent moves, with the spread between Libor and OIS for the middle of next year projected to narrow only to 34bps.

Indeed, there are reasons to think that the reduction in prime MMF assets will have a lasting and material impact on the rate at which banks can fund themselves in CP markets. With few attractive alternatives for non-US banks (especially Japanese lenders) to access the US dollar funding they require – FX forward markets are even more punitive – we are unlikely to see an immediate return to very low funding spreads.

It is likely that even though the reform date has now passed, the impact of prime MMF outflows will remain an important story in months ahead. Indeed, the outflow from prime MMFs in the past year has been matched almost one-for-one with inflows into the government MMF universe, with a US$1,024bn inflow into US Treasury and agency MMFs in the same period that prime funds have seen a US$1,034bn outflow, according to iMoneyNet and JPMorgan.

The ongoing impact of this large asset switch may have significant and lasting implications on the relative clearing price of other short end interest rates with respect to Libor, which in turn could impact the yields on government debt securities relative to interest rate swaps. Looking ahead, money-market dynamics will probably remain very much in focus. 


Next article:
The cross-currency basis: a primer

06 Dec 2016

The cross-currency basis may be one of the most poorly understood concepts in financial markets.

Read More
09 Apr 2019 | By Nicholas Wall
The Cycle May Be Slowly Turning

Better China and European PMI data, the reduced chance of a no-deal Brexit and cooling Sino-US tensions all help to suggest that the economic cycl...

16 Aug 2017 | By Huw Davies
Everything looks expensive!

Over the last 12 months I’ve heard this phrase in conversations with colleagues and investors more times than I care to mention.