— Paul Amery's Blog


On the face of it, mutual funds are a huge global success story.

Assets in collective investment schemes have grown seven-fold in two decades and now top $30 trillion.

Growth in global mutual fund assets

mutual fund assets

Source: Investment Company Institute

Yet the mutual fund structure contains weaknesses.

First, funds need to apportion correctly the costs associated with investors entering and departing the fund, a tricky job. So-called dilution levies (also called entry fees or front-end loads) and exit fees are there to make sure that the costs of any purchases (or sales) of securities incurred on behalf of an entering (exiting) investor are borne by that investor, rather than by the rest of the fund holders.

Unfortunately, many fund managers have historically set entry/exit fees too high and used them as a disguised source of revenue.

A more important drawback of funds is that they may carry liquidity risk.

If a fund invests in less liquid assets but promises immediate liquidity to its investors, the fund manager may cope with a sudden withdrawal of investors’ funds by selling the most liquid of the underlying assets. This would disadvantage those remaining in the fund by leaving them with a rump of less liquid securities. As a result, investors are incentivised to “run” from a fund in stressed market conditions.

Given the increasingly large size of many mutual funds (including ETFs) and the potential liquidity problems in certain asset classes (like corporate debt) it’s not surprising that regulators have been focusing on fund liquidity as a potential source of broader, systemic risk.

In most jurisdictions, the managers of mutual funds have the little-advertised ability to “gate” (ration) redemptions—a right that, if invoked, could contribute to a panic.

Mutual funds have historically been expensive, too. ETFs are a welcome recent exception to the trend of overcharging, but fund fees pay for hefty salaries at asset management firms, as well as for the services of a whole range of middlemen: lawyers, accountants, custodians and transfer agents.

The world’s largest investors don’t use mutual funds: they insist on their own, flexible managed accounts. What if smaller investors could follow suit?

It turns out that they now can.

Firms like US-based Motif Investing allow you to buy a diversified, thematic portfolio of stocks in one transaction. Motif charges $9.95 for a trade in up to 30 stocks, much cheaper than online brokers’ rates for buying or selling 30 individual securities. You can customise your own portfolio (called a motif) or copy someone else’s.

You end up with an investment that’s very similar to an index fund, but with significant advantages: no annual fees, no administrative overheads, no worries about how costs are mutualised and no concerns over gating. Unlike some funds, a stock basket doesn’t lend shares, use derivatives or incur collateral risks.

Under this model you own the stocks in your motif, with full entitlement to dividends. You retain voting rights, rather than delegating them to an anonymous fund manager, who may not even use them (except, potentially, if the small size of your basket means you own fractional, rather than whole units of shares).

Mutual funds first came into being as a way of spreading risk. Reducing company-specific exposure via a shared portfolio seemed an attractive prospect for European investors recently scarred by Tulipmania and the South Sea Bubble.

But that technological advance came nearly three centuries ago. Now we can see the prospects for another significant shift in the savings market. The emergence of cheap, tradeable thematic baskets of stocks threatens to render funds obsolete.

Read More

Active fund managers may offer poor value for money. But some index fund costs are hidden too.

Professor David Blake of Cass Business School has just released a paper in which he argues for better disclosure of the costs of active management.

Between 50-85% of the costs of active management fall outside the headline “ongoing charges” figure that every European retail fund has to display, estimates Blake.

Commissions and taxes paid on securities trades undertaken by the fund manager are excluded from ongoing charges, for a start.

Other, harder-to-measure “invisible” costs, as Blake calls them, include the bid-offer spreads payable on trades in securities, transaction costs in any underlying funds, hidden revenue earned by the fund manager from stock lending or interest, market impact costs and market timing costs.

The high undisclosed costs of active fund management are often used by advocates of passive funds as an extra justification for indexing.

Unfortunately, some of the arguments being made in favour of indexing are fallacious.

In another study published this month, Blake and co-authors Tristan Caulfield, Christos Ioannidis and Ian Tonk argue that active management is not worth paying for.

Having measured the performance of 561 domestic UK equity funds against the FTSE All-Share index over 10 years, Blake, Caulfield, Ioannidis and Tonk conclude that “the average mutual fund manager cannot ‘beat the market’ (i.e., cannot beat a buy-and-hold strategy invested in the market index), once all costs and fees have been taken into account.”

There’s a major problem with this argument. Tracker funds don’t match the market index, as Blake appears to assume as a matter of fact: they underperform as well.

Clearly, you should expect any index-tracking fund or ETF to lag its index by the annual ongoing charges, which are admittedly smaller on average than those of active funds.

But in harder-to-track equity markets the underperformance of the average index fund/ETF may be greater than the headline fee would imply. This occurs as the result of extra, frictional costs that closely resemble the hidden charges Blake talks about in the context of active management.

For an example of this effect, note that most trackers of the MSCI Emerging Markets index underperform it by more than their ongoing charges (see this recent Lyxor publication for evidence).

This additional performance lag occurs as a result of access costs: you pay more in the form of bid-offer spreads, taxes and commissions when buying stocks in smaller and less developed equity markets. Index funds have lower turnover, on average, than active funds, but they do trade.

[Disclosure—I’ve recently done writing work for Lyxor]

Index fund managers can, however, mask their funds’ underperformance in ways that are even less transparent than some of the below-the-surface active fund costs described by Blake.

One way to do this is to include extra costs in the index (whether as an index fee or, more cleverly, by depressing the assumed post-tax dividend rate in a total return index to make the fund’s performance look better).

If the index fund manager also gets to produce the index, a practice that the US regulator now appears to be condoning, then there’s even more scope and incentive to embed additional costs, to the investor’s detriment.

Another way of hiding index fund costs is to include collateral in a derivatives-based fund that is of lower quality (from the perspective of the repo market) than the quality of the securities basket represented by the index. This generates extra income for the counterparties to the repo trade, who can pocket the revenue with the average investor none the wiser that he’s losing out at all.

The equivalent practice in a physically replicated tracker fund is to exchange the index securities via a securities loan for lower-quality collateral, again resulting in extra income, a large portion of which has historically been retained by the fund issuer or its affiliate.

There’s a little more transparency regarding these practices than there used to be, and more of this hidden income is now being credited to fund investors than before. But securities lending and repo market trades are also much more characteristic of index-tracking than active funds, meriting extra scrutiny when indexing is involved.

It’s worth reiterating that if you buy an index fund assuming that you’ll get the index’s return, you are making a mistake. And if you compare index funds on the basis of their headline fees only, you are ignoring potentially substantial hidden costs.

I’m a fan of ETFs and am all for criticising active fund managers if they are providing poor value for money. But we shouldn’t give indexing a free ride as a result.

Read More

According to the Financial Times (subscription required) the US central bank is concerned about the potential for a run on retail bond funds. The Fed has discussed imposing mandatory exit fees on those holding funds as a way of managing this run risk.

The FT reminds us that US retail investors have pumped over $1 trillion into bond funds since early 2009. Central bankers worry about investors buying funds that promise immediate liquidity while investing in potentially illiquid underlying assets like corporate bonds.

Ironically, the increasingly desperate search for yield by income-deprived investors is squarely the result of central banks’ own zero interest rate and quantitative easing (QE) policies. In recent months there’s been a top-of-the-market feel about bond ETF inflows, for example, with buyers focusing on the highest-yielding (and junkiest) part of the market, like leveraged loan funds.

I wrote in 2012 that regulators were soon likely to focus on fund liquidity risk. There’s an inherent tension between the promise of the retail fund “wrapper”—to allow investors to enter and depart at will—and the underlying assets of some funds.

In Europe, fund exit fees already exist in a rough and ready way in the form of redemption gates in fund prospectuses. If more than 10% of investors in a retail (UCITS) fund want to leave in a single day, the fund administrator can postpone redemptions until the next day.

Interestingly, some ETF issuers have told me that they’ve waived this right in the past, deciding not to halt occasional daily redemptions exceeding 10% of fund assets, presumably for fear of negative publicity.

ETFs, which are traded intraday on stock exchanges, provide a pressure valve in the sense that secondary market trading in a fund with suspended redemptions might continue—but it would be in limited volumes and presumably at a big discount to the fund’s net asset value.

We saw this happen on a small scale during a mini-panic in US-listed corporate bond ETFs last June, with one market-maker stepping away from its promise to process investors’ sell orders.

In the US mutual fund market, suspending redemptions is trickier because of the liquidity promise built into the 1940 Act structure. Fund managers must meet redemption requests within seven days and a suspension of redemptions is possible only in extreme circumstances.

At least 85% of fund assets have to be invested in “liquid securities” (I’m not sure how leveraged loan ETFs, which comply with the 1940 Act, can claim to be doing that).

You can see why the Fed is worried about run risk in junk bond mutual funds, particularly since there’s evidence that US investors are flightier than their European counterparts.

By cracking down on the banks, which can no longer hold large inventories of bonds for trading purposes, regulators have shifted risk to bond mutual funds. Now they are concerned about the consequences of their own policies.

I’ve written in the past about the potential for a crash in this sector. In particular, the shaky dealing infrastructure underlying bond funds seemed (and still seems) to me like an accident waiting to happen.

All the same, I feel uneasy about regulators’ current proposals to intervene so directly in the fund market. Investors buy funds knowing they may lose money. You can’t outlaw market bubbles and busts: they happen. If central bankers really think they can manage away all systemic risk then the next crash will be all the more violent.

Read More

Manipulating interest rate, currency and commodity benchmarks will soon be a criminal offence in the UK, carrying the threat of a seven-year jail sentence, according to press reports.

Meanwhile, evidence mounts of an epidemic of benchmark-rigging.

The 2012 LIBOR scandal has been followed by other alleged cases of price-fixing, involving gold and foreign exchange.

Traders have gone on the record to describe traders’ attempts to skew survey-based energy benchmarks by submitting unrealistic prices that suited their trading positions—just as occurred with LIBOR.

Other types of market manipulation appear to have taken place in the financial instruments used by indices underlying tracker products.

In 2009 I published an article describing how investors in oil ETFs were being hit by apparent front-running in the futures contracts being used by the trackers as reference prices.

Unsurprisingly, the business of producing indices and benchmarks is now under intensive regulatory scrutiny.

Those compiling benchmarks have to comply with IOSCO’s 2013 principles for financial benchmarks, which set new governance and transparency requirements, by next month. The European Commission has its own draft benchmark regulation on the table.

Some of the firms involved in benchmark-setting are deciding it’s not worth the hassle—or the increasing potential legal liability—to continue. Deutsche Bank recently resigned from the daily London gold and silver fixes, having failed to find a buyer for its seat on the fixing panel.

Requiring greater transparency in benchmark-setting, introducing new sanctions and bringing about a greater separation between those calculating benchmarks and those involved in trading should help to counter the worst abuses.

But none of these things will make benchmarks immune from attempted manipulation.

Even share indices, which, unlike oil, FX and interest rate benchmarks, are calculated using prices from relatively transparent public markets, are not fool-proof.

All indices undergo periodic rebalancing. When index providers announce details of the stocks due to enter and exit their benchmarks some traders seek to profit from the information during the period between the announcement and the date of the index change itself.

Index firms can change the benchmark rules to try and counter attempted front-running, for example by randomising the dates on which the rebalancing occurs within that time-frame. This may deter manipulators, but it also detracts from the index’s transparency. A benchmark using opaque construction rules wouldn’t work as a benchmark at all.

There’s no perfect solution here. The most popular benchmarks are used in financial products because they have gained widespread acceptance and met a market need. $350 trillion of derivative products were reportedly linked to LIBOR, which was originally seen as a superior, market-based alternative to official central bank interest rates. But such volumes of traffic attract those seeking to influence prices or flows.

The more accepted an index or benchmark becomes, the greater the inefficiencies it may generate.

Heavier fines and jail sentences may deter the worst cases of benchmark-fixing. But some attempts at market manipulation are an inevitable result of the use of indices and benchmarks as reference prices in financial products.

Read More

Regulators are angry about those abusing benchmarks.

This week the head of the International Organisation for Securities Commissions (IOSCO), David Wright, said that benchmark riggers should face tougher sanctions.

Wright’s comments come in the wake of a fine handed down by the UK securities regulator, the Financial Conduct Authority (FCA), to Barclays for lax controls in allowing a trader to influence the daily gold fix in 2012.

The trader’s motivation was to prevent his bank having to make a payout under a derivative contract called a digital option. The trader succeeded in influencing the fix to his benefit, but the client who had bought the derivative suspected manipulation and complained. Barclays, together with the FCA, unveiled the fraud.

This abuse is directly reminiscent of the LIBOR manipulations revealed in the same year, 2012. According to one hedge fund manager quoted in yesterday’s Financial Timesthe influencing by banks of  prices in the markets underlying such digital options is “routine” and the Barclays case is “the tip of the iceberg”.

The average investor can draw one obvious conclusion from the latest episode of benchmark manipulation: don’t buy a financial product from an institution with the motivation and the capacity to influence the reference price underlying the product.

The Chinese walls supposedly in place to separate the benchmark calculators and the traders within a single organisation may have ears.

That applies just as much to the market for index-tracking ETFs and ETNs, where some product issuers also design and calculate the underlying index. In the wake of the LIBOR and gold fixing scandals, such a lack of separation of duties is a warning.

The problem in regulating all of this is that some of those closely connected to current public policymaking don’t have clean hands.

Government agencies (in the form of central banks) are arguably the biggest benchmark riggers of all: think attempts to control foreign exchange and interest rates and the current policy of quantitative easing, which has driven bond yields to well below and asset prices to well above fair value.

And central banks may have been involved in worse behaviour than that. A commenter on yesterday’s Financial Times article, Julian Wiseman, former head of interest rate strategy at bank Société Générale, makes the observation that:

“The most prolific offenders at defending digital barriers (binaries, knock-ins, knock-outs) used to be the Asian central banks. Around the turn of the millennium a significant part of the FX market was knowing who wanted what. Investment banks would sell the exotic options to Asian CBs knowing that they would lose on that position, but that loss was merely the price of the information, which could itself be used profitably.”

Will IOSCO and other regulators now prosecute the central banks, Wiseman asks, rhetorically?

Uncovering this type of market abuse will be harder than hitting the banks while their reputations are at a low ebb.

But if regulators are serious about fixing the system of market benchmarks they should broaden their investigations to include the official sector.

Read More

Asset managers are up in arms about regulators’ possible designation of certain firms or funds as too big to fail.

Any move to classify individual fund managers or funds as systemically important would be likely to impose onerous new capital or liquidity constraints on the entities concerned.

By comparison with banks, which have to hold substantial capital in the form of equity or retained earnings against their balance sheet loans, asset managers operate with a relative sliver of capital backing. This makes them highly profitable. But does it also make them risky?

Asset managers’ standard defence to regulators rests on three arguments: most of us don’t use leverage; we don’t guarantee the value of our funds; and, unlike banks, who act as principals, we are agents who don’t put our own balance sheets at risk.

There are some exceptions to these statements. Hedge funds—who are asset managers—are leveraged. Constant NAV money market funds in the US have long operated with something akin to a promoter guarantee of stable value. And asset managers’ expanding activity in the secured finance (repo and securities lending) markets may incur balance sheet risks for the firms involved.

As banks’ operations are crimped by new regulation, asset managers are reportedly taking up the slack in other, riskier areas of the financial markets too, like prime brokerage, trading and leveraged lending.

But for all their current lobbying dollars, asset managers may be misunderstanding regulators’ main concern—excessive scale.

The 2008/09 crisis exposed great vulnerabilities in the structure of the financial system. Its near-failure ended up costing hundreds of billions of taxpayer dollars.

From the perspective of network theory, the single riskiest type of actor in a network is one that operates at scale and with multiple links to other players.

The failure of such a “high-degree node”—one with many links—can easily bring down the whole system.

In the financial system, regulators can take action to reduce the risk of a key player failing by forcing bigger firms to hold more capital, downsize or even break up.

By comparison with banks, asset managers’ overall significance in the financial network is open to debate.

But you can see that regulators may be concerned about the huge scale of some asset management firms, and that of some of the funds they run.  Many large mutual funds promise instant liquidity while investing in less-than-liquid assets.

The investors in those funds should be well aware that the value of their holdings may fluctuate. But are they prepared for a logjam at the exit door if everyone tries to get out at once?

Add to that asset managers’ growing involvement in traditional investment banking activities and regulators are likely to err on the side of caution in classifying them. If an asset manager were at the root of the next crisis and regulators hadn’t acted, imagine the furore. Remember AIG, an insurance company, which blindsided them last time?

For this reason the largest asset managers, and the biggest funds they run, may well face new regulatory constraints.

And for BlackRock, Vanguard, Pimco, Fidelity and State Street, the largest asset managers, restructuring and reducing scale, perhaps by the demerger of different business activities, may be the way to avoid painful new capital and liquidity requirements.

Read More