On Tuesday, February 25th Bloomberg carried an article outlining a proposed quarterly asset tax on large financial institutions. The proposal is expected to be unveiled this week as part of Dave Camp’s (Republican chairman of the House Ways and Means Committee) tax reform proposals. We take a look at the proposal’s impact on repo.
The source of the article was “a Republican aide with knowledge of the plan” who “described it on condition of anonymity.” The tax would impact those institutions which hold assets in excess of $500 billion. Currently this would include 10 institutions including banks – JP Morgan Chase, Bank of America, Citigroup, Wells Fargo & Co. Goldman Sachs and Morgan Stanley – as well as GE, AIG, Prudential and MetLife. A charge of 3.5 basis points would be assessed quarterly against any assets in excess of $500 billion. The balance sheet calculation would be taken from the 2010 Dodd Frank legislation. The fees would however be deductible against the institution’s federal income tax liability.
The rational for this new tax is that it will allow the federal government to reduce corporate and individual tax rates while keeping the overall tax code revenue neutral. It is anticipated that the tax would raise $86.4 billion over the next decade. The overall impact on these institutions is unclear at this point. These institutions could find that the additional taxes are offset by the reduction in their marginal tax rates. However, it should be pointed out that the asset tax will be assessed regardless of profitability and will have a disproportionate impact on low margin parts of these institutions’ business. Also, major global institutions with significant operations in the United States would not be impacted.
From a political point of view we find this proposal somewhat surprising. Even if it is revenue neutral for a majority of the institutions affected, it is somewhat troubling that the federal government may be starting down the path of industry specific taxation. A similar tax (a financial crisis responsibility fee) was proposed by the Obama administration in the wake of the financial crisis. Under that proposal a fee would have been assessed against balance sheets in excess of $50 billion but adjustments would be allowed for less risky assets. The Obama proposal would have only remained in effect until the cost of TARP was recovered. Camp’s plan would become permanent and would actually raise $27 billion more over the next decade than the President’s proposal.
But putting politics aside for a minute, we can foresee a disproportionate impact on the secured finance market as the proposed tax would impact all bank assets equally regardless of return. Repo and securities lending have traditionally been significant users of balance sheet but generally minor users of capital; they are balance sheet intensive but low margin. Much of the reform which has been proposed to date has been designed with the intention of reducing systemic risk, and by assessing additional capital to risky transactions various regulators have ‘encouraged’ banks to decrease their exposure in these areas. This has often been offset by granting banks favorable treatment for risk reducing strategies such as central counterparty clearing. And this in turn has allowed banks to manage their match book business in a traditional manner albeit on a reduced scale. Over the last month we have seen the Basel Committee reverse it earlier stand on repo netting and we have noted a change in sentiment towards the proposed pan European FTT.
Until this article surfaced we felt that the most important piece of proposed regulation still under consideration was the suggestion of a resolution authority here in the United States. This proposal is now a potential game changer for repo desks. Keep in mind that this is not being undertaken in the name of regulation and risk reduction. It is a change in the tax code and its impact is easy to assess. SIFMA’s website reports an average of $1.769 billion of reverse repos on the books of the primary dealer network during the month of January. Ignoring the possibility of netting that would equate to a fee of $619 million per quarter or almost $2.5 billion per year! Put another way, matchbooks would need to generate a spread of 14 basis points over the course of the year just to break even. While we cannot assess the odds on this proposal becoming law, we can predict that if it were the ability of dealers to fund client assets would be lost.