VTB - Plan to Buy Back Preferred Shares

VTB's CFO commented earlier this week that the bank is in talks with the Finance Ministry to potentially reduce the proportion of preferred shares in its total equity. The details are not finalized, but if there is an agreement to do this, we expect that VTB would look to do it in a "capital-neutral" way by using funds that would be paid as preferred share dividends to buy back a portion of the preferred shares each year, most likely over a several-year time frame. Over time, this would imply a transfer of value from preferred shareholders to common shareholders, and we think a cleanup of the capital structure would be positive for VTB's investment case. We note, however, that VTB's capital position is not particularly comfortable, so it would seem to us that the process would indeed take several years and be contingent upon VTB's profitability improving and it being able to generate some capital internally. On a nearer-term note, VTB's CFO guided net income of R70-75 bln this year, which would imply around 5% ROE and highlights the bank's challenging profitability backdrop. > Why is this being proposed? The thinking is that VTB's complex capital structure has long been confusing and negative for the stock's investment case, albeit neutral for VTB debtholders, and this proposal would work toward cleaning it up. VTB currently has three types of shares outstanding: common shares (12.96 trln, nominal value R129.6 bln); Type 1 prefs (owned by the Finance Ministry; 21.4 trln, nominal value R214 bln); and Type 2 prefs (owned by the DIA; 3.1 trln, nominal value R307 bln). So R521 bln in nominal pref shares are outstanding, and the prefs currently account for a large part of the bank's CET1 regulatory capital (R1.3 trln as of September 1). VTB also has $2.25 bln in perpetual debt with a 9.5% coupon included in its AT1 capital.> How would it work? Nothing has been finalized, but as far as we understand, the provisional plan is that instead of paying dividends to pref shareholders, as is currently the case (based on equalizing yields with the common shares), the portion of profit allocated to the prefs would instead be used to buy back pref shares. The current thinking is that in order for this process to be "capital-neutral," it would extend over several years. We understand that conversion of the pref shares to common shares is not being considered. The pref shares that are bought back would be held in treasury and potentially cancelled.> What impact would this have on VTB's capital? As is fairly well known, VTB's capital ratios have been quite thin for a while and have not been helped by fairly weak profitability. As of end-1H20, its consolidated ratios were: CET1 (N20.1) - 9.3% (minimum including capital buffers: 8%); Tier 1 (N20.2) - 10.2% (minimum: 9.5%); and total CAR (N20.0) - 11.8% (minimum: 11.5%). VTB's most recent standalone capital ratios as of September 1 were: CET1 (N1.1) - 9.1%; T1 (N1.2) - 10.1%; and total CAR (N1.0) - 11.4%. At this stage, VTB is operating with a quite tight total CAR, partly because the capital contribution of its legacy T2 capital instruments ($1.5 bln T2 Eurobonds due in 2022) is amortizing, so the bank plans to substitute these instruments, which are phasing out, on the local bond market, which we think will be possible only gradually given the size of the local market for such instruments. Buying back all of the preferred shares would imply a hit of around 360 bps to the standalone CET1 capital, and the bank clearly does not have a sufficient CET1 capital cushion to implement this currently. Even if profitability and internal capital generation pick up materially next year, we still think this looks like a process that will take several years. > What's in it for the Finance Ministry? Over time, buying back the preferred shares would imply a transfer of value (earnings streams) from the preferred to the common shares, or from the state to common shareholders. Obviously, the Finance Ministry is the largest common shareholder (with a 60.9% stake), so it would benefit directly from this transfer as well as from any positive improvement in the investment case. Moreover, these preferred shares were issued in 2014-15 as part of "temporary" support measures for the banking system, so (as has been seen globally) their repayment can be seen as repaying this temporary support. It could also be seen to amount to a form of "quasi-privatization," which we assume would be viewed positively in the government. > How would it affect our equity valuation? We value VTB using capital and earnings attributable to common shareholders, so over the next several years, it would not have any material impact on the valuation, and we don't even have the final details. We also have concerns that this would weigh further on VTB's fairly fragile capital position, but if it can be made to work, we think cleaning up the capital structure could support a reduced discount rate for VTB over time. > Near-term earnings outlook looks weak. VTB's CFO also mentioned that VTB expects to earn around R70-75 bln in net income in 2020 (we model R85 bln), which implies R20-25 bln in profit in the final four months of the year, with provisioning expected to be high in 3Q20 and real estate write-downs high in 4Q20. This would imply around 5% ROE and highlights the challenges VTB faces to materially boost earnings and capital generation (recall that the last two annual dividend payouts have been cut from an intended 50% to 15% and 10%, respectively). > Implications for bondholders. For holders of VTB's subordinated bonds, we think any plans to buy back shares would be perceived as negative, given the bank's typically relatively tight capital ratios, and a commitment to implement the buyback would likely restrain VTB from potentially building up its capital cushion over regulatory minimums in the near future. Should the bank manage to implement the buyback very gradually and in a capital-neutral way, this would not change the current state of things much. Furthermore, it is important that VTB's outstanding $2.25 bln perp issue has a write-down trigger linked to the bank's total CAR (the threshold is 7.5%), rather than the CET1 ratio, and the bank clearly has more flexibility in terms of managing its total CAR (e.g. through raising T2/AT1 capital on the local market).