July 27, 2017
Ms. Stephanie L. Sullivan
Senior Technical & Policy Advisor
Office of Financial Services
Division of Corporate Finance
Securities and Exchange Commission
100 F Street N.E.
Washington, D.C. 20549
| Re: | Shinhan Financial Group Co., Ltd. |
| | Form20-F for Fiscal Year Ended December 31, 2016 |
| | Form6-K Filed May 15, 2017 |
Dear Ms. Sullivan:
We are writing in response to your letter, dated June 30, 2017, containing the comments of the Staff of the Securities and Exchange Commission (the “Commission”) on the annual report on Form20-F with respect to the fiscal year ended December 31, 2016 filed with the Commission on April 28, 2017 and the current report on Form6-K filed with the Commission on May 15, 2017.
Our responses to the Staff’s comments are set forth in this letter and each response follows the text of the paragraph of the comment letter to which it relates.
***********
Form20-F for the Year Ended December 31, 2016
Note 3 - Significant Accounting Policies, pageF-14
(ab) New Standards and Interpretations Not Yet Adopted, pageF-34
1. | We note your disclosure on pageF-35 that the actual impact of adopting IFRS 9 on the financial statements in 2018 is being assessed and is subject to change because it will be dependent on the financial instruments that the Group will hold and economic conditions at the time as well as accounting policy election and judgments that you will make in the future. However, we are also aware of statements made by your CFO on the February 8, 2017 earnings call (per review of the edited transcript) that indicates that you expect IFRS 9 to result in a positive for the capital ratio, and that the loan loss reserves of the present were already converted in 2016. Please respond to the following: |
| • | | Please explain in further detail what is meant by the statement that the “loan loss reserves of the present were already converted in 2016.” Specifically, please clarify whether in your December 31, 2016 financial statements prepared in accordance with IFRS as issued by the IASB, you have made adjustments to your methodologies in contemplation of complying with IFRS 9 early. If so, please describe the changes made and adopted, along with a quantitative impact of the changes. |
| | | | |
Securities and Exchange Commission | | 2 | | July 27, 2017 |
Response:
A third party provided the English transcript of the February 8, 2017 earnings call, which reflects such third party’s English translation of a transcript originally prepared in Korean.
During the earnings call, it was the Managing Director of the Finance Management Team (“Managing Director”), and not the CFO, who spoke about IFRS 9. The Managing Director made a general statement that the adoption of IFRS 9 would have a significant impact on the financial industry.
As described on pageF-139 of our Form20-F (Note 30(g)), Shinhan Financial Group (“the Group”) has been accumulating regulatory reserves for loan loss in accordance with the regulations of the Financial Supervisory Service applicable to Korean financial institutions. The Managing Director mentioned during the call that, as a result of amendments to relevant regulations in 2016, beginning from the end of 2016, regulatory reserve for loan loss has been recognized as capital for the purposes of calculating BIS capital adequacy ratio. The Managing Director made a statement to the effect that such recognition of regulatory reserve for loan loss as capital since the end of 2016 has had a positive effect on the Group’s BIS capital adequacy ratio.
In sum, the statement that the “loan loss reserves of the present were already converted in 2016” means that when calculating the BIS capital adequacy ratio in 2016, the Group’s regulatory reserves for loan loss were recognized as capital; the statement does not mean that the Group adjusted its financial statements in accordance with IFRS 9 at the end of 2016.
| • | | Please explain to us in further detail the impact you expect upon the adoption of IFRS 9. For example, it appears from the statements made by your CFO on the earnings call that you expect a positive capital ratio impact, but please also clarify whether you expect a positive impact on IFRS shareholders’ equity (and reduction in the allowance for loan losses) upon the adoption. As part of your response, address what the main drivers are for the positive impact, if applicable, and whether it is concentrated in certain loan portfolios or a specific measurement method, e.g. the12-month expected credit losses. |
| | | | |
Securities and Exchange Commission | | 3 | | July 27, 2017 |
Response:
As further described in our response to the bullet point below, the Group is still in the process of evaluating the expected effects of adopting IFRS 9.
| • | | In future filings, given that you appear to have some estimates of the impact of the adoption of IFRS 9, please provide more quantitative information about the expected effect upon the adoption of the standard on January 1, 2018. |
Response:
The Group is currently under the process of evaluating the expected effects of adopting IFRS 9, including searching for appropriate methodologies to calculate such expected effects by performing simulations using various criteria. The Group plans to disclose in future filings the impact of the adoption of IFRS 9 on the financials once the Group finalizes its IFRS 9 accounting policies.
Note 26 – Employee Benefits, pageF-123
(e) Sensitivity Analysis, pageF-125
2. | We note that your sensitivity analysis showing a 1% increase and 1% decrease in the discount rate results in a 92% increase in the defined benefit obligation, or a 110% decrease, respectively. We note similar material changes in the 1% movement of the future salary increase assumption. To the extent that the sensitivity amounts are driven by the fact that the amounts represent full numbers (and not numbers in millions of Won), please revise future filings to make it clear in all situations where you are not disclosing amounts in millions of Won. |
Response:
The sensitivity analysis table on pageF-125 shows the full amounts of the defined benefit obligations upon the changes contemplated under the sensitivity analysis; the table reflects the total revised obligation rather than the change in obligation resulting from the sensitivity analysis.
For example, in the sensitivity analysis table on pageF-125 of our Form20-F (Note 26(e)), the “Increase” column of the “discount rate (1%p movement)” line item shows that when the discount rate increases by 1%p, the defined obligation becomesW1,542,736 million, not that the defined obligation changes byW1,542,736 million. Therefore, millions of Won is the correct unit in the sensitivity analysis table.
| | | | |
Securities and Exchange Commission | | 4 | | July 27, 2017 |
However, in the future, in order to avoid confusion, the Group will report the increase/(decrease) in the defined benefit obligations that result from the 1%p fluctuations in the discount and the future salary increase rates rather than the total revised defined benefit obligation.
Form6-K Filed May 15, 2017
Exhibit 99.2
Note 2 – Basis of Preparation, page 14
(b) Use of Estimates and Judgments, page 14
3. | We note your disclosure that the credit card segment revised the calculation methodology of loan losses from a roll rate analysis model to an internal model approach. As a result of this change, allowances for loan losses and provision for unused credit lines decreased byW81,226 million andW282,694 million, respectively. We note that such changes represent a 12% reduction in the allowance for loan losses for credit cards and a 65% reduction in the provision for unused credit lines compared to the balances as of December 31, 2016. Please respond to the following: |
| • | | Please explain in more detail the driver(s) behind this change in methodology during the first quarter of 2017. As part of your response, please clarify whether this is the same approach you plan to utilize upon your adoption of IFRS 9. If so, please confirm that this methodology is still based upon the guidance of IAS 39, and not IFRS 9, given that you do not plan to adopt IFRS 9 until January 1, 2018. |
Response:
BASEL II requires a minimum of nine years of data collection (consisting of a minimum five-year observation period for defaults and a minimum four-year observation period for post-default recoveries) as a necessary condition to using the internal model approach. After its merger with LG Card in 2007, Shinhan Card has worked to establish a risk management system and met the BASEL II nine-year data collection requirement in October 2016.
At the end of December 2016, the Financial Supervisory Service granted Shinhan Card final approval to use the internal model approach. During the first quarter of 2017, Shinhan Card completed the establishment of the IFRS loan loss calculation system, for example, by replacing BASEL II risk components with risk components for financial reporting in accordance with IAS 39.
| | | | |
Securities and Exchange Commission | | 5 | | July 27, 2017 |
The main reasons the Group’s credit card segment (“Shinhan Card”) revised its methodology for calculating loan losses in the first quarter of 2017 are as follows:
The internal model approach allows a more sophisticated calculation of loan loss.
The internal model approach calculates separate default rates and loss given default for different groups of customers, differentiated based on the characteristics of both the customers and the products that they use. The internal model approach disaggregates customers into more than twice the number of groups than does the roll rate model.
Whereas the roll rate model does not distinguish between customers with high and low risks of default when calculating roll rates, the internal model approach allows for a more sophisticated calculation of loan loss that reflects the customers’ credit ratings.
Shinhan Card transitioned to the internal model in accordance with IAS 39 in the first quarter of 2017, resulting in the changes in estimates. Upon the adoption of IFRS 9 in January 1, 2018, Shinhan Card will continue to use the loan loss calculation system based on the internal model approach and reflect new IFRS 9 requirements.
Shinhan Card’s use of the internal model approach improves consistency of the Group’s accounting standards.
The Group aims to maximize consistency of the accounting standards used in its consolidated financial statements, and therefore the Group’s subsidiaries should, to the extent possible, use consistent calculation methodology of loan losses. Shinhan Bank has been applying the internal model approach since 2011. Therefore, Shinhan Card’s use of the internal model approach has improved the consistency of the Group’s accounting standards.
| • | | Please explain to us in more detail why the impact was so much more significant to the provision for unused credit lines, as compared to the allowance for loan losses for credit cards. |
Response:
Adopting the internal model approach during the first quarter of 2017 had a more significant impact on the provision for unused credit lines than on the allowance for loan losses for credit cards because the internal model approach, unlike the roll rate model, reflects the differences in customer credit ratings when calculating loan losses and unused credit lines and credit cards reflect different distributions of customer credit ratings.
Approximately 85% of the unused credit line exposure is to customers with high credit ratings. Customers with high credit ratings have low default rates and infrequently use cash advances, such as card loans.
| | | | |
Securities and Exchange Commission | | 6 | | July 27, 2017 |
As the internal model approach applies default rates that reflect customers’ credit ratings, the provision for unused credit lines under the internal model approach reflects the high proportion of customers with high credit ratings. As a result, the internal model approach yields a much lower provision for unused credit lines than does the roll rate model, which does not take into account the customers’ credit ratings and instead applies a single roll rate table for each product.
On the other hand, the credit card customers are not as concentrated among those with high credit ratings and instead show a more even distribution among those with varying credit ratings. Therefore, the use of internal model approach has a less significant effect on the allowance for loan losses for credit cards than on the provision for unused credit lines.
| • | | Please tell us any changes you made in 2017 to your internal controls as a result of this change in methodology for your credit card segment. |
Response:
In connection with its changes to the loan loss calculation methodology for financial reporting purposes in the first quarter of 2017, Shinhan Card also replaced the internal controls related to the use of the roll rate model with the following internal controls related to the use of the internal model approach:
| • | | The risk management team conducts self-verification and subsequently seeks internal approval of its yearly calculations of risk components such as probability of default, loss given default and credit conversion factor. The Group’s modelling verification team and auditing team further review the methodologies and results of the risk component calculations. |
| • | | Each quarter, in order to ensure that the loan loss accurately reflects the appropriate balance and risk components of each asset pool, the risk management team confirms whether the balance for each asset pool used to calculate loan loss matches the balance for such asset pool reflected in the calculation of the Group’s BIS capital adequacy ratio and verifies the loan loss through recalculation. |
| • | | On a yearly basis, the risk management team verifies the appropriateness of the division of asset pools based on probability of default. |
| • | | If the risk management team discovers any unusual trends or discrepancies in monthly trends of unused credit lines, the risk management team reevaluates the appropriateness of the amount of provision for unused credit lines. The financial planning team reflects such amount in the final settlement of accounts after conducting final checks such as comparison against initial estimates. |
| | | | |
Securities and Exchange Commission | | 7 | | July 27, 2017 |
| • | | If the risk management team discovers any discrepancies in loan loss accounting records (including the balance in the system and the loan loss calculations sent to the accounting department) or unusual trends in monthly changes in the loan loss for major products, the risk management team reevaluates the appropriateness of the amount of loan loss, including through an evaluation of the underlying reasons for the changes in the loan loss. The financial planning team reflects such amount in the final settlement of accounts after conducting final checks such as comparison against initial estimates. |
***********
Please do not hesitate to contact myself at+82-2-6360-3071 or our external counsel, Jin Hyuk Park of Simpson Thacher & Bartlett (35th floor, ICBC Tower, 3 Garden Road, Central, Hong Kong, telephone number+852-2514-7665 and fax number+852-2869-7694), if we can be of any assistance to the Staff.
Kindly acknowledge receipt of this letter by stamping the enclosed copy of this letter and returning it to our messenger.
Thank you in advance for your cooperation in connection with this matter.
| | |
Very truly yours, |
|
/s/ Yim Bo-hyuk |
Name: | | Yim Bo-hyuk |
Title: | | Deputy President & |
| | Chief Financial Officer |