WHY SO COMPLICATED?

The more technical a law or a regulation is, the higher the need for industry expertise. Without this expertise, legislators risk making serious mistakes. Thus experts who advise the crafting of norms have a vested interest in making those norms complicated, because that increases the value of their expertise and human capital. It is possible that this is just an unwanted side effect of the experts’ desire for precision. But when you read an op-ed written by the senior officers of one of the largest economic consulting firms asking for more cost-benefit analysis, it is difficult to conclude that it is serendipitous.²

In 2010, we saw the passage of the Dodd-Frank financial-reform bill, which was a staggering 2,319 pages long. Things were not always this way. The Glass-Steagall Act, which in 1933 separated investment banking from commercial banking, was just thirty-seven pages long. The act that created the Federal Reserve in 1913 ran to thirty-one pages. Even the recent Sarbanes-Oxley Act, which was written in response to the Enron and WorldCom scandals, was only sixty-six pages long. Tellingly, the Dodd-Frank bill was popularly called the “Lawyers’ and Consultants’ Full Employment Act of 2010.” It may well have created more jobs than Obama’s original 2009 stimulus package did.

Each page of regulation probably provides a year’s worth of employment for several lobbyists and a couple of lawyers and economists. This gigantic waste is never properly factored into our economic analysis. But the biggest cost is the smokescreen that overregulation creates. For centuries, in Continental Europe, laws were written in Latin, a language that ordinary citizens could not understand. As institutions were democratized, laws started being written in the vernacular. Overabundant regulation, and the legalese that it is written in, achieves the same goal that Latin once did: to confuse the public. “When I was writing regulations,” says one retired EPA regulator, “I was told on more than one occasion to make sure I put in enough loopholes. The purpose of the complexity is to hide the loopholes.”³

Back in 2009, at the beginning of the financial-reform legislative process, I made a concerted effort to follow the debate. It was my area of expertise, so I wanted to know about it, but at a certain point I gave up. Except for a few selected passages, I read a summary of the final Dodd-Frank bill, rather than the full 2,319-page version, to say nothing of the sixty-seven studies that the bill mandated and the thousands of pages of further regulation that it authorized.⁴

During the debate, I was stunned by the popular support for the old Glass-Steagall Act. The more distant people were from the financial world, the stauncher their support for restoring the separation—which had been defunct since 1999—between commercial banking and investment banking. As an economist, I knew that the separation had costs as well as benefits, and in the absence of strong evidence that the benefits exceeded the costs, I went with my free-market assumption that the separation wasn’t a good idea. The 2008 financial crisis did not change my views. While investment banks were among the biggest culprits in the crisis, the merging of commercial and investment banking activities, in my view, was not to blame. Bear Stearns was only an investment bank, not a commercial one, and it nevertheless got a bailout because the Federal Reserve deemed it too interconnected to fail. So what did it matter whether a commercial bank and an investment bank were jointly owned or simply joined at the hip by a web of financial contracts? Further, commercial bank JP-Morgan Chase subsequently bought Bear Stearns, reducing the public cost of intervention—another sign, I thought, that a forced separation between the two forms of banking was unwise.

So why was the 1999 repeal of Glass-Steagall so unpopular? Indeed, why is it still so unpopular that the Occupy movement demands that the separation be restored? For a long time, I thought they just didn’t get it: I was an expert, and I knew best. But eventually I realized that I was the one who didn’t get it. While Glass-Steagall may not be the most efficient form of regulation, it worked for more than sixty years. People sensed that the power of the financial industry had become excessive, and they wanted to contain it. They might not understand the sophisticated economic arguments, but they understood the bottom line: that the best is the enemy of the good, and that by trying to achieve the best possible regulations, we end up preventing the passage of feasible regulations.

The beauty of Glass-Steagall, after all, was its simplicity: banks should not gamble with government-insured money. Even a six-year-old can understand that, which is why former Fed chairman Paul Volcker endorsed the principle in 2009. But then the Obama administration transformed the so-called Volcker rule into 298 pages of mumbo jumbo.⁵ Why? Well, as an economist, I have been trained to infer intentions from outcomes, especially when I’m considering smart and experienced people. So it seems a reasonably safe bet that the Obama administration endorsed the Volcker rule, rather than Glass-Steagall, because it knew that the Volcker rule was almost as popular but, unlike Glass-Steagall, required so many details to be implemented that it would never be enforced. In this way, the Obama administration could avoid displeasing the banking industry and get political consensus to boot.

The United States was born on the principle of no taxation without representation. It should add no regulation without representation. But if regulation is too complex, people have no way to understand it and thus cannot participate properly in democracy. Simplifying regulation, therefore, is essential to building a capitalism for the people.

SIMPLE IS BEAUTIFUL

Simple regulation is necessarily inefficient, at least in a narrow economic sense, as Glass-Steagall was. So why impose simplicity in spite of this inefficiency?

When I visited Stanford Business School many years ago, I was surprised to see that all of the offices in its new building were identical—a result that had cost money, thanks to the structure of the building. Why should socialism prevail with respect to offices? I was told that the dean, who had to assign the offices, wanted to avoid the headache of having to decide who would get the best ones. Faculty members’ compensation is never disclosed, but office size is obvious. We professors are all extremely concerned about our prestige. If some offices were visibly more desirable than others, then each professor would lobby to get the best spots. The dean wanted to prevent that.

At the time, I thought these concerns were exaggerated, until Chicago Booth also constructed a new building for itself but decided to differentiate offices. To minimize lobbying, the dean announced that each faculty member would be randomly assigned a number within categories—presumably assistant, associate, full, and chair professor (though this was not explicit), and would choose an office sequentially. But when the selection order was announced, the most famous faculty members were first, suggesting that the process had not in fact been random. The school erupted. Emotions took over. One faculty member shouted “I hate you!” at another who had received a better office, ruining their relationship for quite some time. Some faculty members organized a simulation to try to figure out how this sequential procedure would have assigned offices. For days, the only topic of conversation was office assignments. Ironically, except for a few corner offices, the offices didn’t differ that much. Most of the haggling was over nothing.

We might underestimate the cost of all this commotion because it was not easily measurable. But if you do factor in the time wasted in office-allocation simulations, along with the cost of tense relationships for years to come, you see that Stanford’s choice was the more efficient one. This point has been recognized by a few economists.⁶ When you factor in the enforcement costs and lobbying costs of regulation, many choices that looked inefficient at first become efficient in a broader sense. So when you simplify regulation, you might make it less effective in curbing the distortion you might want to curb, but you actually reduce its overall cost and make it more transparent. By ignoring enforcement and lobbying costs, policy makers fall all too often into the “Nirvana fallacy” of comparing real world markets with ideal regulation. Simple regulation stands a better chance to allow a comparison between real-world markets and real-world regulation.

In the context of regulation, however, there is one added benefit of simplicity. Not only does simplifying regulation reduce lobbying costs and distortions; it also makes it easier for the public to monitor things, reducing the amount of capture.

If you still fear the inefficiency of simple regulation, let me make one final argument. It is often useful to make an instrument inefficient when you fear that it will be used too much. Conservatives, for instance, oppose the introduction of a value-added tax because it raises revenue so efficiently, making it very easy for politicians to expand the size of government. The same logic applies here. Regulation that is a bit inefficient by design will reduce the overuse of regulation. Only the most compelling and useful regulation will survive.

SIMPLE ENFORCEMENT

One of the many benefits of simple rules is that they facilitate accountability. Complicated rules are difficult to enforce even under the best circumstances, and impossible when their enforcement is the domain of captured agencies.

Between the Federal Reserve, the Office of the Control of the Currency, and the State Regulators there are about 30,000 people in charge of supervising banks. Given this army, how is it possible that bank supervision is so ineffective? In March 2008 the market was predicting the probability that Washington Mutual would fail within a year, at 30 percent. Still the bank was left to operate until September 25 of that year. There is no political payoff for an early intervention, especially given the uncertainty that surrounds all such decisions. After the government took over Washington Mutual in 2008, there were still complaints that the action came too early. Preventive banking regulation is like pre-emptive war: There is no credit for the pain avoided, while there is plenty of blame for the pain inflicted. Experience shows that we simply can’t rely on regulators to resist these pressures.

The alternative, however, is not to do away with regulation but to rely on simpler regulations that have other enforcement mechanisms. In Chapter 14 I will present a proposal on how to use the market to regulate banks without relying on ineffective banking supervisors. Similar ideas can also be applied to other forms of regulation. As I explained in Chapter 4, whistleblowers identify corporate fraud more effectively than the SEC does and at a fraction of the cost. And as noted above, a key advantage of a whistleblower-based system is that it is resistant to capture. Any employee can be a whistleblower, and it is too costly for the industry to buy them all off. Thus, antifraud regulation can more effectively be enforced by amply rewarded whistleblowers.⁷

THE SIMPLICITY OF ZERO

When it comes to enforcing rules, zero tolerance is often the only policy that makes sense: zero tolerance for crime in the street; zero tolerance for sexual harassment; zero tolerance for exceptions in a promotion system. Zero tolerance makes no allowance for trade-offs and therefore prevents an erosion of standards.

If a police officer observes a small criminal act on the street, should he intervene? Of course. But if he is allowed to use his own discretion, he will likely intervene to a lesser degree than the rule requires. Since he bears the personal and physical cost of intervening, consciously or subconsciously he will lean a bit too much against acting. Other officers observing his behavior might assume that his decision reflects the effective rule. When confronted by their own decisions, they will likely copy his leniency. Little by little the standards are eroded, unless a zero-tolerance policy dispenses with ambiguity.

We may recognize the damage that subsidies and loopholes do to a market economy, but there is resistance to adopting a zero-tolerance policy toward them. In isolation many of them seem worthy: a tax reduction to promote investments in the inner city; subsidized loans to help develop green energy and save the environment; a reduction in payroll taxes for young unemployed people. The problem is that when you open the floodgate there is no realistic restraint. Each industry, each large corporation, employs an army of lobbyists to pressure Congress on their merits. They flood newspapers with op-eds, some paid for directly, others indirectly. They advertise on television and start “grassroots” campaigns on the Internet. And who is lobbying for the other side? Nobody.

AN APPLICATION TO THE PERSONAL TAX CODE

How can we force regulation to be simple? With 60,000 pages and 3.4 million words, the tax code is the obvious place to begin. There are millions of justifications for exemptions and deductions. Individually, they may even make sense. But collectively, they do not. The deductions that make it into the code are likely to be those that benefit the most politically powerful groups. The existence of some deductions encourages everybody else to lobby for similar deductions, as in the case of the business-school office assignments. If there is a home-mortgage deduction, why not a deduction for motor homes? What about boats, cars, bicycles? The only defensible line is zero. The moment the door is opened to one deduction, we have a flood, and resources will be wasted on lobbying for deductions, as well as on the lawyers and tax accountants who exploit them.

Paradoxically, many of these deductions lead to results opposing their goals. Consider the mortgage-interest tax deduction, which defenders claim encourages homeownership and the attainment of the American Dream. The fact that the deduction is unlimited implies that it subsidizes the wealthy, who have large mortgages and consequently can take large deductions. In cities where real estate is limited—such as New York, San Francisco, and Boston—the subsidy will have the effect of raising home prices. Since the increase will be in proportion to the average subsidy, the less wealthy will wind up receiving less than the cost of home increases, jeopardizing the American Dream. Meanwhile, the subsidy induces excessive investment in housing and excessive leverage within the household sector, a problem that emerged after the 2008 financial crisis.

Further, the proliferation of tax deductions erodes the tax base and forces marginal tax rates to increase. As I will explain in the next chapter, higher marginal tax rates encourage people to evade and elude taxes. The most extreme form of tax elusion is substituting leisure for work. When I watch a movie, I do not get taxed; when I work, I do. Taxing income is equivalent to subsidizing leisure. The more I am taxed on work, the less work I do.

To get a sense of how deductions make marginal tax rates increase, I took 2008 tax returns and calculated what the marginal tax rates would have been if we had eliminated all deductions and raised the same amount of revenue. The result: people with household incomes below $30,000 would have been exempted from any tax. The fraction of income between $30,000 and $70,000 would have been taxed at 10 percent. The remaining marginal tax rates would have been 20 percent, up to $150,000; 25 percent, from $150,000 to $250,000; 30 percent, from $250,000 and $500,000; and 35 percent above that.⁸ In short, getting rid of all deductions would have reduced the marginal tax rate of all taxpayers by at least five percentage points, except for taxpayers making more than $500,000, who would see their marginal tax rate unchanged.

Far from being regressive, such a reform would decrease the tax burden of all households making less than $1 million a year. The additional tax burden for those in the income bracket between $1 million and $1.5 million would be $22,000 a year—probably close to what they currently spend on lawyers and tax accountants, who would be mostly unnecessary if we got rid of all deductions.

I do not claim that the rates used in my simulation are optimal. My point is simply that we can simplify the tax system without making it more regressive. The majority of proposals advocating simplification have suggested a flat tax rate, an idea that I will discuss in the next chapter. Here, suffice it to note that simplifying the tax code would be a good thing. The only argument against it is a Machiavellian one: tax loopholes make taxing less efficient and consequently reduce the risk that government will increase the tax rates. For this reason, it would be fair to introduce a constitutional amendment imposing a maximum marginal tax rate, something I will also discuss in the next chapter.

SIMPLIFYING CORPORATE TAXES

One change that would greatly simplify the tax code and avoid plenty of tax arbitrages is the equal treatment of personal income and capital gains. However, reducing the personal tax rate to the level of the capital-gains tax rate would create a large hole in the US budget, while increasing the capital-gains tax rate to the level of the personal income tax would discourage investments. Fortunately, this problem can be easily resolved by modifying the corporate tax rate. Currently, a dollar of profits is taxed 35 cents at the corporate level, and then another 9.75 cents (15 percent tax on the remaining 65 cents) when it is distributed in the form of a dividend or realized as a capital gain.

The problem with such a high corporate tax rate is that it is very distortive and all too frequently eluded. It’s distortive because partnerships are heavily advantaged vis-à-vis corporations and so is debt, whose payments are deducted from the taxable corporate income. Also, large corporations with armies of lobbyists are very good at eluding this tax, while small firms have to pay it. A 2008 study by the General Accounting Office found that 55 percent of large US companies paid zero taxes at least one year between 1998 and 2005.⁹ What’s the impact of that? Well, in 1962 corporate tax revenues amounted to 39 percent of corporate profits and 16 percent of total tax revenues. In 2007, they represented only 29 percent of corporate profits and 10 percent of tax revenues.¹⁰ This elusion is not the same in every sector. A Bloomberg Business Week analysis, based on public filings, infers that on average semiconductor companies pay only 19.6 percent of their income in taxes, while telecommunication companies pay only 22.2 percent.¹¹

One time-honored way companies reduce their tax liability is by hiring lobbyists to obtain special tax loopholes or to protect the ones they have. For example, in 2003 Tyco paid lobbyist Jack Abramoff’s firm $150,000 a month to fight legislation aimed at the company, which reincorporated in Bermuda to avoid paying US corporate taxes.¹² The proposed legislation would have cost the company about $4 billion. Tyco’s strategy succeeded. Another example of businesses buying preferential tax treatment is the dividend repatriation provision in the American Jobs Creation Act of 2004, which allowed US multinationals a one-time opportunity to bring home foreign earnings, paying taxes on only 15 percent of this repatriated income. A study finds that the ninety-three public firms engaging in lobbying for this provision spent $282.7 million on their persuasive effort, and won $62.5 billion in tax savings, with a 220:1 return on investment.¹³

The simplest solution to all these problems is to move most of the burden from the corporate level to the individual level. In fact, corporations have much stronger incentives to lobby for tax loopholes than individuals do. If corporate taxes were just 15 percent and dividends and capital gains were taxed at the personal tax rate (35 percent), a dollar of corporate profits would still yield the same tax contribution before reaching shareholders, but tax elusion and tax arbitrages would be more difficult.¹⁴

Elimination of the deductibility of interest from the definition of taxable corporate income could easily help lower the corporate tax rate to 10 percent without reducing tax revenues. Such provision would eliminate the distortions in favor of debt present in the current tax code, which disadvantages small firms (which are unable to borrow in the same proportion as large ones), and pushes all firms to borrow more.

THE SIMPLICITY OF CERTAINTY

At the end of 2010, there was enormous uncertainty about future taxes: were the Bush-era income tax cuts to expire, as planned, at the end of 2010, tax rates would increase for everybody—at the lower end of the income distribution from 10 to 15 percent, and at the higher end from 35 to 39.6 percent.

The greatest uncertainty, however, was over estate taxes. By 2010, they had been reduced to zero. Absent congressional action that year, they were to leap to 55 percent for estates valued above one million dollars in 2011. The change was so abrupt that a doctor expressed the fear that unscrupulous people might accelerate the death of sick rich relatives. ¹⁵ The estate tax reprieve was granted at the eleventh hour, but the income tax cuts were extended only for another two years, setting the stage for another last-minute fight in 2012.

Regardless of the merit of these tax cuts, one has to despise their temporary status: it generates great uncertainty, which is bad for the economy as well as for individuals (not to mention the old sick relatives). Unfortunately, the Bush tax cuts are not an exception. An increasing number of provisions in the tax code are designed to expire. A recent study shows that in 1991 there were about 40 expiring tax cuts, but in 2011 there were close to 400.¹⁶ Why did the number of these measures increase ten times in twenty years?

One (not particularly noble) reason is that to federal bean counters, expiring tax cuts appear smaller than permanent ones. The Congressional Budget Office (CBO) computes its projections for federal spending and borrowing using ‘current law’ as a baseline. Thus, if a tax cut is set to expire in a given year, the CBO assumes that the expiration date will trump politics and the tax will be reinstated. This rule allows the CBO to whistle past the graveyard and score tax cuts in a way that makes them appear more affordable.

Another reason I can think of is even less noble: expiring tax provisions are a bonanza for lobbyists and congressional fund raisers. The threat of an impending tax increase or an expiring tax loophole can mobilize clients and raise more funds. This is the vicious circle of lobbying. Companies hire hoards of lobbyists to obtain special treatment. While the lobbyists desire a reputation for success, they do not want to be so successful that they are no longer needed. It would be in the interest of lobbyists to create a dependence, which expiring tax provisions do very nicely. For this reason, tax rules should not only be simple but also permanent.

SIMPLE AND EQUAL

The best way to ensure that rules are simple and effective is to apply them to Congress, where they originate. Seems like common sense? Until February 2012, members of Congress were not bound by the same insider-trading restrictions that applied to corporate insiders. After several suspicious trades by top members of Congress were documented in a book, this anomaly has been quickly rectified, at least on paper.¹⁷ It remains to be seen how aggressive would be the Securities and Exchange Commission to go after Senators and Congress people. But the problem goes well beyond insider trading.

The US Congress lives by rules that are very different from those imposed on ordinary businesses. If the corporate executive of a publicly traded company lies during a conference call with analysts or in an annual report, he can be sued. Politicians, on the other hand, lie regularly during electoral campaigns and while in office, with no consequences. Wouldn’t it be nice if we could sue a lying politician?

If the US government had been held to the same accounting rules as the private sector, it would have been forced to consolidate Fannie Mae and Freddie Mac and to report all contingent liabilities at market value.

Not only would such a rule make government officials appreciate the costs of the constraints they put on others, but it would greatly improve government’s accountability. One of the reasons why credit subsidies are so appealing is that they are not immediately accounted for, so they are invisible to most of the electorate. The cost of the implicit support provided to Fannie Mae and Freddie Mac did not appear on the government books for many years. This made it easier for the government to continue supporting them.

The same is true for the support provided to large banks and to students. In principle, this distortion could be partially eliminated by reforming the way the cost is accounted for (or not accounted for). Yet any cost that is uncertain and will occur in the future permits some accounting discretion. There is enormous pressure to hide these costs, and the temptation to fudge the rules is very large. The only way to rout out dangerous accounting gimmicks when calculating cost to the taxpayers is by ensuring that the government follows all the rules that apply to publicly traded companies, including the personal liability of the Chief Executive Officers and the Chief Financial Officers for the integrity of the accounting numbers.

ENFORCEMENT TO THE PEOPLE

No matter what we do to try to level the playing field, it will remain unbalanced. Businesses will always want to lobby Washington for favors. Even if there were a law against government subsidies to any company or industry, clever lawyers and lobbyists would find ways around it. But suppose we introduced such a law along with an enforcement mechanism based on class-action lawsuits—which, for all their abuses, are very important to American democracy. Such a mechanism would grant any citizen the right to sue the subsidized industry in the name of the state and obtain restitution of the unfair subsidy, retaining a finder’s fee—say, 15 or 20 percent—for himself. The system would be modeled after qui tam actions, in which a private individual who assists a prosecution can receive a portion of any penalty imposed. It could with equal effect be used against the power of lobbies.

A reward system for whistleblowers who help expose corruption can also be an effective way to replace ineffective government agencies. As I explained in Chapter 4, whistleblowers identify corporate fraud more effectively than the SEC does and at a fraction of the cost. And as noted above, a key advantage of a whistleblower-based system is that it is resistant to capture. Any employee can be a whistleblower, and it is too costly for the industry to buy them all off.

CONCLUSIONS

In this chapter, I have argued that to overcome capture we need limited and simple regulation, preferably enforced by a whistleblower reward system. The best way to prevent lobbying is to introduce a law eliminating subsidies to industry, supported by giving citizens the right to sue to recover the cost of subsides paid unfairly. But would these measures reduce the power of government to the extent that it becomes completely ineffective? No. As I will argue in the next chapter, everything that can be done with subsidies can be done with taxes as well. In fact, it can be done better, because the political economy of taxes is more favorable to ordinary citizens.