## Andrew G Haldane: Turning the red tape tide
Remarks by Mr Andrew G Haldane, Executive Director, Financial Stability, Bank of England, at the International Financial Law Review Dinner, London, 10 April 2013.
* * *
Regulatory and legal frameworks share common roots. Both are complex, evolutionary systems, shaped by history. They are the result of a set of well-intentioned historical actions by technicians charged with filling cracks, creating certainty, shaping incentives for the common good. Both legal and regulatory frameworks have many of the characteristics of a classic public good.
But the cumulative consequences of even well-intentioned actions may not always deliver outcomes which necessarily serve society well. That is because such actions are typically a response to events and circumstance. The resulting frameworks have a history of pathdependence (David (1985)). This history-dependence may 'lock-in' sub-optimal technologies, such as QWERTY keyboards and VHS video-recorders.
The financial regulatory framework may be another example of a locked-in technology. The events which have shaped the evolution of financial regulation include crises, scandals, innovation and liberalisation. Each has elicited a response, typically the addition of a new regulatory layer. The cumulative consequence has been a regulatory tide which has tended to flow in only one direction - towards a lengthier, more complex rulebook.
In this respect, financial regulation shares many evolutionary characteristics with elements of the legal framework - for example, the tax code. It too has been responsive to events and circumstance including crises, scandals, innovation and liberalisation. And for many of the same reasons, it too has resulted in a rising tide of rulebook complexity. The facts are striking.
In the UK, the Banking Act (1979) covered 52 sections and 75 pages. By the mid-1980s, the Financial Services Act (1986) and an updated Banking Act (1987) had expanded primary legislation to 110 sections (106 pages) and 212 sections (299 pages) respectively. The Financial Services and Markets Act (2000) took this to 433 sections or 321 pages. And the new Financial Services Act (2012) takes this to 695 sections or 534 pages - a tenfold increase in primary legislation in a generation.
In the US, the picture is no less dramatic. Contrast the legislative responses in the two largest financial crises of the past century - the Great Depression and the Great Recession. The Great Depression spawned the Glass-Steagall Act (1933) - perhaps the single most important piece of financial legislation of the 20 th century. That ran to a mere 37 pages. More recently, the Great Recession has spawned the Dodd-Frank Act (2010). It runs to 848 pages or more than 20 Glass-Steagalls. Once completed, Dodd-Frank might run to 30,000 pages of rulemaking.
The tax code has followed a similar evolutionary trajectory. In the UK in 1997, Tolley's Tax Guide ran to around 5,000 pages. By 2009, it had reached over 11,500 pages; just two years later, 17,795 pages. That is a threefold rise in less than 15 years.
In the US, the Federal tax system is celebrating its centenary. In 1913, US Federal tax rules summed to 400 pages. By 1945, this had risen to 8,200 pages, by 1974 to 19,500 pages and by 2000 to 46,900 pages. In 2013, US Federal tax rules amounted to 74,000 pages. That is almost a two hundred-fold increase over a century.
Viewed through the lens of history, it is not difficult to explain this evolution. As cracks have emerged in the plaster - for example, regulatory and tax loopholes - they have been papered over. Like a bad painter and decorator, new layers have been applied to support the
layers below. These were rational responses to real problems. And for a time they will have filled cracks, reduced uncertainties, corrected incentives.
Yet this evolutionary path leads inexorably towards a framework - whether regulatory or tax which is a patchwork of make-do-and-mend. History locks in the idiosyncrasies and complexities of the past, generating a steadily rising tide of red tape. As the regulatory and tax codes attest, the resulting frameworks can be Byzantine in their complexity and HeathRobinson in their design. For three distinct reasons, this patchwork may not serve society well.
First, maintaining such a framework is a high-cost activity. In financial regulation, it makes for a steadily-rising standing army of regulators and compliance officers. In the US in the 1930s, there was roughly one regulator for every three banks. Today, there are roughly three regulators for every US bank. In Europe, the cost of implementing Basel III is estimated at over 70,000 full-time private sector jobs (Harle et al (2010)). In the US, the cost of DoddFrank is estimated at tens of thousands of jobs. As these resources might otherwise have been profitably deployed in other industries, these are deadweight opportunity costs borne by society.
The direct and opportunity costs of the tax system are no less great. The Taxpayer Advocate Service recently estimated that US households and companies spend as much as 6.1 billion hours per year in filling out tax returns. Others have estimated the annual cost of tax compliance in the US at around 1% of annual GDP (GAO (2005)) - or about $160 billion in 2012. Similar estimates are found in Europe.
Second, complex frameworks tend ultimately not to solve the problems for which they were a response. They may be ineffective. In filling old cracks, complex rulebooks tend to open up opportunities for new ones to emerge. Indeed, they may increase the likelihood of new loopholes or workarounds emerging. Using complex rules to control a complex environment carries the risk of chasing one's tail in ever-decreasing circles (Gigerenzer (2007)).
It is not difficult to spot this tail-chasing problem in the financial regulation sphere. For example, the shift to a highly-complex risk-weighting system for bank assets has resulted in the average amount of capital banks assign to these assets halving over the past 20 years. Why? Because complex rules have generated both the incentives and the means to exploit regulatory loopholes. That is why simple measures of bank leverage, untainted by such complexity, were ten times better at predicting banking failure during the crisis than complex regulatory alternatives (Haldane and Madouros (2012)).
What is true of financial regulation is true too of tax. For example, studies have analysed the incidence of tax evasion and avoidance across different countries. They have found that the single most important determinant of tax evasion is often the complexity of the tax code (Richardson (2006)). The greater the complexity, the more numerous the loopholes, the greater the incentive and means to exploit them.
Third, complex regulatory frameworks tend also to be inequitable. They advantage those best able to exploit the cracks, navigate the uncertainty, squeeze through the loopholes. This tends to be those with the deepest pockets who can afford the most sophisticated riskmodeller, the slickest tax accountant. Complexity, in other words, acts like a regressive tax.
That is why big banks (using in-house complex models) typically hold far less capital than smaller banks (using simplified, standardised alternatives) even when their underlying exposures are identical. It is one reason why Warren Buffet claimed to pay a lower tax rate than his secretary. And it is why, over the recent past, regulatory and tax arbitrage have both become high-growth industries. They are toxic exhaust fumes from the Heath-Robinson vehicles history has created.
If complex frameworks come with economic and social costs - inefficient, ineffective and inequitable - why has society not done more to tackle them? Resistance is strong, particularly among those who gain most from squeezing through the loopholes. There is also an in-built professional inertia among regulators, lawyers and tax accountants with large
amounts of human capital invested in complexity. In removing complexities, society's gain would be their loss.
Nonetheless, in the light of the financial crisis, we may be nearing an inflection point, where the societal pendulum begins to swing in the opposite direction. There are certainly signs of the red tape tide beginning to turn in the area of financial regulation. A growing chorus of concern has emerged recently about complexities and inconsistencies in banking regulation. Earlier this year, the G20 Finance Ministers and Central Bank Governors committed themselves to addressing this problem. So too has the Basel Committee on Banking Supervision.
This is a good start. But, at this stage, it is no more than that. One risk is that the regulatory inclination may be to tackle this problem slowly and piecemeal, to unwrap the onion one layer at a time, to avoid throwing the baby out with the bathwater. As Axel Leijonhufvud once remarked to me in a different context, while the bathwater in this case may be worth reusing, the baby would never have come of anything.
Thin-slicing reform, whether regulation or tax, condemns us to failure. Peeling the onion one layer at a time tends to end in tears. It is only by removing all of the old layers that the painter and decorator can achieve a robust finish. Removing these one by one will simply reveal old cracks, open up old loopholes. Tax reform has been successful when it has recognised that fact and has started simple and afresh. Regulatory reform is no different.
For financial regulation, that means asking some big questions. Is risk-weighting of bank assets worth the costs? Are bank's internal risk models more regulatory trouble than they are worth? Does complexity unduly advantage large incumbents over small new entrants? Are armies of supervisors and compliance officers a sign of success or failure? These questions should form the new regulatory battleground.
This could transform the regulatory framework, for the simpler, for the better. It could do the same for taxation. Some will argue that we face a Canute-like task in turning this tide - and, like Canute, that it will be in vain. But, unlike the seas, this complex tide is ours to turn. With the economy struggling for breath, the time has never been better.
## References
David, P (1985), 'Clio and the Economics of QWERTY.' American Economic Review , 77(2): 332-337.
Gigerenzer, G (2007), 'Gut Feelings: Short cuts to Better Decision Making', Allen Lane .
Government Accountability Office (2005), 'Summary of Estimates of the Costs of the Federal Tax System', available online at http://www.gao.gov/new.items/d05878.pdf.
Haldane, A and Madouros, V (2012), 'The Dog and the Frisbee', paper presented at the Federal Reserve Bank of Kansas City's 36th economic policy symposium.
Härle et al (2010), 'Basel III and European banking: Its impact, how banks might respond, and the challenges of implementation' McKinsey & Company.
Richardson, G (2006), 'Determinants of tax evasion: A cross-country investigation', Journal of International Accounting, Auditing and Taxation 15 (2006) 150-169.