Speaker: Mary Starks, Director of Competition and Economics
Location: Social Market Foundation (SMF) lecture, London
Delivered on: 27 February 2017
- Fixing or capping prices can have very serious unintended consequences.
- Policy makers should review price caps regularly to make sure they work as intended.
- Price regulation is a public policy decision that can create both winners and losers.
Note: This is the speech as drafted and may differ from the delivered version.
Let me take you back to the times of the Emperor Diocletian in the Roman Empire, around the end of the third century AD. Diocletian brought in a number of reforms that changed Roman society for the better: changing the structure of the government and creating more equitable taxation. He also, had the radical idea to curb inflation via price controls.
Perhaps the ancient equivalent of announcing a significant policy reform at midnight, on Twitter.
Diocletian’s edict on maximum prices tried to contain the ‘frenzied avarice’ of ‘unscrupulous’ traders by declaring maximum prices for around 1,000 retail items across the empire, disregarding any considerations of supply and demand, or transportation costs.
As you can imagine, it was not long before black markets flourished. After a short and ineffective period, the edict was abolished.
My point is that for at least 1700 years, controlling prices has appealed to policy makers, but has proved tricky in practice.
I often get asked whether the FCA is a ‘price regulator’. It’s a deceptively straightforward question.
I’d love to give a simple yes or no answer and finish the speech here, but the answer is it depends (I am an economist after all).
Unlike some other regulators (Ofcom, Ofgem and the other utility regulators), we do not have formal, regular price reviews built into our remit. In this sense, we are not a price regulator.
We can regulate price in some situations. The important question is: when should we?
Our remit is to promote competition in the interests of consumers, to safeguard market integrity, and to secure an appropriate degree of protection for consumers. In advancing these objectives, there have been some instances where the FCA has intervened to set a cap on certain prices charged to consumers. So clearly we can regulate price in some situations. The important question is: when should we?
I will discuss this in three parts today. First I’ll look at some other examples of successful and less successful price regulation, and draw some lessons. I’ll then talk about the cases in which FCA does control prices directly. Thirdly, I will talk more broadly about why FCA cares about price, and how we think about intervening. This can be indirectly – to shape the forces that bear upon price, or directly – to set, cap or otherwise control price.
The topic is both technically complex and politically charged. For the last 30 years or so, since the privatisation and liberalisation of the Thatcher years, the strong orthodoxy in economic policy circles has been that price control comes in only when there is no hope of getting competition to do the job – that competition is always preferable to price control. However, that orthodoxy is now under challenge from across the political spectrum, and we have seen the re-emergence of price controls for pre-payment customers in energy, and a lively debate about extending caps further.
Lessons from rent controls
Turning to another page of history, let’s jump forward to the Second World War, when President Franklin D Roosevelt introduced the Emergency Price Control Act to counter the inflationary pressures of full wartime employment.
One of the consequences of the Act was to freeze New York rents at their 1943 levels. Despite the normalisation of the economy following the end of the war, the Act was replaced with new laws that retained price controls on rents, and there have been variations of these controls since then.
While this has helped keep rents low for people in rent-controlled flats, it has had the opposite effect for everyone else.
Forcing rents below the market level has led to landlords withdrawing housing stock, and reduced the incentive to build more housing. This has contributed to housing shortages, pushing non-controlled rents up. As in Diocletian’s time, these price controls also led to market forces expressing themselves in other forms, such as landlords taking under the table payments or instituting high sign-up fees.
While rent control appeals as a way to curb landlords’ profitmaking at tenants’ expense, most economists agree that the unintended consequences seriously undermine the goal of affordable housing. In fact, Paul Krugman has said that rent control is among the best-understood issues in all of economics, and—among economists, anyway – one of the least controversial. This is not to say rent control can never work but the unintended consequences have to be recognised and addressed.
So why is the policy still in place, more than 70 years after the end of the war? And why did the current mayor, Bill de Blasio, campaign for rent freezes in 2013? The policy appeals to tenants, promising lower prices, greater certainty, and maintaining the city’s character, and to unwind the policy could be seen as appeasing landlords and punishing tenants.
Fixing or capping prices can have very serious unintended consequences, including undermining the very intent of the policy and fuelling black market or other criminal activity.
I want to draw two lessons from history before we move on to financial services and the present day. First, fixing or capping prices can have very serious unintended consequences, including undermining the very intent of the policy and fuelling black market or other criminal activity. Secondly, once a policy is in place, even if it proves to be bad policy, it can become entrenched – the direct beneficiaries are deeply vested, while those who lose out are harmed indirectly and invisibly. The politics runs counter to the economics.
Both Diocletian and FDR deployed price control rein in inflation. These days we deal with inflation through other means, and most policy discussions about price control are microeconomic – they are about prices for particular goods and services.
The obvious example of this is in utilities, where price control has been used to curb monopoly power. The track record of price regulation in utilities, in the UK and internationally, has been well studied and could fill an entire lecture series.. I’m going to move right on to financial services.
Price caps and controls in financial services
Financial services infrastructure – notably the payments system and securities trading infrastructure (exchanges, clearing houses etc.) – can have elements of monopoly power and network effects. There are various complex, arcane and in some cases disputed issues I could go into here, but I’ll pick just one – interchange fees.
If you are a competition economist or lawyer, you will know what these are because you will at some point in your career have been paid to understand them. If you are a normal person, the interchange fee is one of the most ubiquitous prices you have never seen. When you pay with plastic for, say, a loaf of bread, the baker’s bank pays your bank an interchange fee. The fee is set by the card scheme, Mastercard or Visa. And for reasons that a competition economist or lawyer can explain to you over the least enjoyable pint of beer of your life, competition tends to drive these fees up not down.
Much of the developed world tried to tackle interchange fees using competition law, which was fiendishly complicated and thus extremely lucrative for competition advisers, even if it didn’t do much for their pub banter.
However, Australia took a different tack. The Reserve Bank of Australia set up a Board with the power to cap interchange fees for credit cards. Since 2006 this has been capped at 0.50 per cent of the value of the transaction for both Visa and Mastercard.
When these caps were proposed, some firms cautioned that it would crush the credit card industry. Reducing the amount of revenue firms could earn through interchange fees would cut margins and force firms out of the market.
However, two subsequent reviews by the Reserve Bank of Australia, in 2008 and 2015, have confirmed that the cap appears to be addressing the distortions in the market without critically impairing the credit card industry. Similar regimes have subsequently been introduced in the US (in 2011) and the EU (in 2013).
So price control has proved more effective than competition, in this instance. What did the Australians do right? First, they diagnosed the problem correctly (if counter-intuitively) – it was not a cartel-like problem for which competition law was well suited; it was a market power problem to which price capping is well suited. Second, they committed to reviewing the policy at regular intervals. As it happened, the policy did not have the unintended consequence of crushing the credit card industry, but if it had they would have spotted it early.
Price regulation in the context of utilities or financial infrastructure is primarily motivated by concerns about monopoly or market power.
Price regulation in the context of utilities or financial infrastructure is primarily motivated by concerns about monopoly or market power – companies that enjoy a privileged and unassailable position, and whose ability to over-charge their customers must therefore be reined in.
But most financial services markets don’t look like that. Most financial services markets have multiple players. Most regulation of retail financial services is primarily motivated by the complex, long-term and high stakes nature of the products. If you buy a lousy loaf of bread, you go to a different baker next time.
So why has price regulation become ‘a thing’ in these markets?
Price regulation by the FCA
In recent years the FCA has introduced price regulation in three areas: on high-cost short-term credit in 2014, workplace personal pension schemes in 2015, and early exit pension charges in 2016.
In each of these cases the duty to regulate prices was placed on the FCA by the government. So while, in theory, we are able to initiate price regulation using our own powers, we have not done so since our inception in 2013. This is significant. As the New York rent control example illustrates, there are often winners and losers from price control. Policies that create losers as well as winners involve difficult political choices as well as technocratic design challenges.
In the high-cost short-term credit market, we imposed a set of caps on the fees and charges that can be levied on borrowers: a daily interest cap of 0.8%, a default fees cap of £15, and a total cost cap of 100% of the principal. We imposed a range of other measures at the same time.
Why cap the price of high-cost short-term credit? There were lots of lenders in the market at the time. But competition on price was weak – borrowers were not shopping around or comparing prices, partly because this was difficult, but also because they wanted credit urgently. Borrowers who needed cash immediately were less concerned with what credit cost, and were lent money without much discipline around whether they could afford to borrow. Many then faced excessive fees and charges when they couldn’t repay, suffered egregious collection practices, and found themselves trapped in debt spirals.
You’ll recall that one of the unintended consequences of Diocletian’s price edict was to drive activity into the black market. One of the main debates around the high-cost short-term credit price cap was whether it would have a similar effect, driving borrowers into the arms of illegal money lenders.
Following in the footsteps of the Reserve Bank of Australia, we recently reviewed the impact of the cap. We found that 60% of people rejected for high-cost short-term credit did not go on to borrow from other sources. The rest mainly went to friends or family, with 15% using other forms of formal credit.
What else did the review tell us? The market is now smaller. In 2016, 800,000 people took out 3.6 million loans to the value of £1.1 billion. That might sound a lot but it is less than half the size it was in 2013. But, vitally, consumers are paying less for loans and are better able to repay them on time.
For workplace personal pension schemes, we imposed, among other measures, a fee cap of 0.75% for default schemes used by firms to meet the automatic enrolment duty. We also prohibited firms from price discrimination based on whether their members currently contribute to the funds. This example is interesting for three reasons. First, it combines price control and nudge – those customers who do not make an active choice find themselves in a price-capped default scheme contributing at the auto-enrolment rate. However firms are free to develop and market other options at other price points but they can only win customers via active choice.
Secondly, we took steps to protect a particular group that might be even less engaged, and without an employer looking out for their interests – former employees who remain scheme members but are no longer contributing. We banned firms from charging these customers more than they charge active members.
Thirdly, and incredibly important, the Government established a supplier of last resort: NEST. By the way, I use supplier of last resort in the policy wonk sense, I am not trying to be rude about NEST – far from it. This protects against one of the most serious unintended consequences of price capping – that some customers become unattractive or even uneconomic to serve. In the pensions market, which typically charges ad valorem fees, these are customers with small pots. There are quite a lot of these, particularly as we are in the early days of auto-enrolment.
Lastly, for contract-based personal pensions, we introduced an early exit charges cap of 1% of the pot. There was a political imperative here to ensure people did not have to pay hefty fees to access the new pension freedoms, a flagship government policy.
We can and do cap prices. But this is the exception not the rule. Across the financial services sector there are tens of thousands of prices set every day by market forces where we do not intervene directly.
But the case is also interesting as an example of capping what Which? has called ‘sneaky fees and charges‘. This is distinct from capping the headline price. One argument for capping less visible fees and charges is that it prevents a competitive dynamic taking hold where there is fierce competition on the headline price, but firms look to recover cost through other fees and charges, including those buried in the small print. I’m sure many of you will recognise this phenomenon from airlines or car hire.
So we can and do cap prices. But this is the exception not the rule. Across the financial services sector there are tens of thousands of prices set everyday by market forces where we do not intervene directly.
How we might use price regulation in the future
So in what circumstances could you make a sensible case for intervening directly to regulate price?
As I mentioned, prevailing orthodoxy was, until recently, that price control was an absolute last resort, only to be tried if all other attempts to boost competition had failed. I basically stand by that.
However, what has become apparent more recently is that competition can drive uneven benefits, and (to paraphrase the old Heineken ads), there are parts of some markets that competition cannot reach. In these cases competition works powerfully to drive prices down for some, but this can be at the expense of higher prices for others. Typically, savvy bargain hunters (or, as we call them in financial services, Martin Lewis readers) do very well, while customers who are less active or ‘engaged’ can do quite badly. Where the goods and services in question are in some sense essential (which is not the case for fashionable restaurants but is the case for many financial services), this undermines confidence in markets as a mechanism for delivery. And where the less engaged are also in some sense vulnerable (for example on low incomes), this is a particularly serious problem.
However, price control is not the only possible policy response. There may be ways to target protection or assistance to vulnerable or at-risk customers specifically. There may be ways to create independent oversight or governance arrangements so that iniquitous pricing practices are challenged and driven out. This was the thinking behind the creation of Independent Governance Committees for workplace pensions, and the governance element of our recent reforms in the asset management sector.
There are, to my mind, two main issues with using price control to protect less ‘engaged’ or ‘savvy’ consumers. The first is the risk of unintended consequences. If you make it less commercially rewarding (or more onerous in compliance terms) to serve vulnerable consumers, fewer firms will want to do it. Having suppliers back away from these customers would be an own goal The second is that unwinding price discrimination creates losers as well as winners, and deciding how far to go requires a view on what is a fair outcome. Regulators tend to prefer that call to be made by elected representatives. Elected representatives may want regulators to be bolder.
Is the FCA a price regulator?
So, is the FCA a price regulator? Well, not in the sense of our colleagues at other regulators who regularly review many of the prices in their remit. We see no reason to intervene directly in most of the thousands of prices that are set every day in financial services markets.
We are concerned when these markets are not working well, for some or all customers, we will generally first look to other policy options to improve the way that competition works. But, clearly there are times when we intervene directly to protect customers, including by controlling prices.
And there will doubtless be further occasions when price control is ‘on the table’ as an option. In those discussions, I will be saying three things.
Simply capping prices can create more losers than winners, and measures intended to help the vulnerable can achieve the opposite.
First, our previous examples have illustrated that price control can mean many, quite different things: from a headline price cap (as with payday lending), to a cap on the price of a default option, (as with auto-enrolment), to a ban on price discrimination between particular groups (as with active and inactive members of a pension scheme), to a cap on small print fees and charges (as with early exit fees). As a crude generalisation, the risk of really serious unintended consequences is probably highest for the first, and lowest for the last.
Which brings me to the second: beware the unintended consequences. The ones I have discussed are restricting supply, making customers – sometimes vulnerable ones – uneconomic to serve, and fuelling black market activity. There are others.
Unintended consequences are not insurmountable. But in designing policy, you need to identify and do something about them. And having implemented the policy, you need to check it is working. Review the impact of the policy thoroughly and soon. And if it is not working as it should, think again.
Third: recognise that these are public policy decisions, with winners and losers. Sometimes that means the final call is rightly the preserve of elected politicians, not appointed regulators.
The ‘frenzied avarice of unscrupulous traders’ makes us angry, and rightly so. But simply capping their prices can create more losers than winners, and measures intended to help the vulnerable can achieve the opposite. Regulators and think-tanks like the SMF need to call this out.