National Bank of Greece (NBGIF) CORRESP Correspondence with SEC

National Bank of Greece

Mr Kevin W. Vaughn

Accounting Branch Chief

United States Securities and Exchange Commission

Washington, D.C. 20549

September 28, 2012

Re:National Bank of Greece S.A.
Form 20-F for the Fiscal Year Ended December 31, 2011
Filed May 15, 2012
File No. 001-14960

Dear Mr. Vaughn:

We refer to your comment letter dated August 10, 2012 on the above referenced Annual Report on Form 20-F (the “20-F”) of National Bank of Greece S.A. (“NBG” or the “Company” or the “Bank”).

Set forth below are the Company’s responses to your comments. For your convenience, the comments are repeated below prior to each response.

Form 20-F for the fiscal year ended December 31, 2011
Item 4. Information about the Company
B. Business Overview, page 44
Retail Banking, page 47
Consumer Lending Products, page 47

1.We note your disclosure on page 48 that you have modified the way in which you grant loans and take on new customers by targeting “customers that plan to invest rather than consume.” Please tell us and revise this disclosure to explain how you determine the intention of your borrowers prior to lending. Please also describe in more detail the specific procedures you take and modifications you have made in your process to ensure you are lending to your target customers. Discuss how this affects the product mix in terms of the types of loans originated as well as the mix within your outstanding loan portfolio.

We advise the Staff that in our next Annual Report on Form 20-F we will revise our disclosure on page 48, as follows:

“… The Bank has also modified the way in which it grants loans and takes on new customers by targeting customers that plan to invest rather than consume by increasing the offer of green loan products (“green loans”) was enhanced through the accomplishment and extension of

various agreements with retail companies specialized in providing innovative technology solutions for energy saving. In addition, the Bank participated in major exhibitions about Environmental Technologies, Energy Saving and Renewable Energy Sources in order to communicate the Bank’s series of green products. We consider that these specific customers clearly have an intention to improve their homes and small businesses instead of using the proceeds of the loans for immediate consumption. The Bank continues participating successfully in the “Energy Saving at Home” program managed by the Institute of Financing Small & Very Small Sized Enterprises (ETEAN S.A., former TEMPE S.A.), supporting the goal for energy efficiency and use of renewable energy sources, which also makes use of a subsidy by a European regional initiative. The proceeds of the green loans are disbursed by the Bank directly to the suppliers either of the services (e.g. building contractors) or the equipment (insulation material, photovoltaic panels etc) and after checking and matching the respective invoice and in this way helping to ensure that the proceeds of the green loans are spent as approved. Demand for green loans is increasing whereas the demand for other types of direct or indirect loans is decreasing. These products represented ·% of disbursements in 2012 compared to 7.6% in 2011 and thus represent a significant part of business activity in 2012 for consumer lending. However, the contribution of these new products in the mix of the total consumer portfolio remains low at 0.2% and ·% at December 31, 2011 and 2012,because the volume of green loan disbursements remains low. ”

Mortgage Lending, page 48

2.We note your disclosure here that the number of housing transactions declined by 44% and house prices by 5.3% in 2011, following previous declines in house prices of 4.4% in 2010 and 4.3% in 2009. You go on to subsequently state that, notwithstanding that information, the housing market is not considered to be overpriced since the decline in house prices over the last three years was rather moderate and gradual and that you expect this trend to continue in the near future. These two statements appear to be contradictory in nature. Please tell us and revise future filings to reconcile this apparent contradiction and to better explain how you are able to assert that you believe the housing market is not overpriced considering the recent continuous year-after-year declines in home prices and the fact that you expect prices to continue to decline into the future.

We advise the Staff that our comment that the housing market was not considered to be overpriced was based on the decline of the ratio of the house price index to the index of rent from 91.0 in 2009 to 84.7 in 2010 and 79.7 in 2011. The moderate decline of house prices since 2009 onwards of 4.3% in 2009, 4.4% in 2010 and 5.3% in 2011 was in line with the perception of house price movements as expressed by the Bank of Greece in their Annual Report for the year 2011 (ed. April 2012, p. 58-60), which states that the decline in house prices during the above mentioned period is considered to be moderate and gradual with no apparent signs of overpricing as measured by the decline in ratio of house price index to the index of rent. Also the moderate decline in house prices was due to lower demand as a result of high ownership rate (over 80%), lower affordability of credit, the fact that family owned construction companies had accumulated wealth over the past years, hence have no immediate incentive to

2

lower prices in order to sell the houses they have constructed, plus uncertainty concerning the employment status, income and future house price evolution. Our view is that the constant decline in prices observed since 2008 is a result of the deterioration in the Greek economy and not a specific bubble or inflationary pressure in the housing market. In particular the decline in prices is in line with the Greek GDP decrease of 2.1% in 2009, 4.2% in 2010 and 6.9% in 2011.

In future filings we will revise this disclosure as follows:

“…Notwithstanding the above, the housing market is not considered to be overpriced as measured by ratio of the house price index to the index of rent, as a result of inflationary pressures or a “housing bubble”. We believe the decline in house prices over the last three years was rather moderate and gradual, and in line with the deterioration in the Greek economy as measured by the decline in the country’s GDP. This trend is expected to continue in the near future following the overall deterioration of the Greek economy.”

3.We note your disclosure on page 50 related to your new “fractional payment” modification program. Please address the following:

·Revise future disclosures to define the term “installment due” as referred to in your current disclosure. Specifically, does the installment due refer to the past due principal, past due interest, or both? Does it include any current amount(s) due?

·Quantify the success rates of this modification program for the period since inception (i.e. mid-2011) through the year ended December 31, 2011, through the quarter ended March 31, 2012 and to the present date. Please tell us and disclose in future filings how many borrowers have been able to timely make payments under this plan versus those that have subsequently re-defaulted.

·In future filings disclose the amount of loans you held at period end that were modified under this “fractional payment” modification plan, and confirm whether these loans are included in the total modified TDR mortgage loans disclosed on page 49.

·Tell us whether loans modified under this “fractional payment” plan are included in the percentage of modified loans which were subsequently modified again as disclosed on page 142. If they are not included, please explain why not.

We advise the Staff that we will revise our future disclosure regarding the “fractional payment” modification program, as follows:

“In mid-2011, as a result of the prolonged economic crisis, the Bank offered to borrowers still experiencing problems in servicing their restructured mortgage, the opportunity to become current in their payments, by offering a new product called “fractional payment”.  According to the terms of the new modified loan (which includes all past due principal and interest, as well as any current amounts due) the borrower pays for a 3-year period a proportion of the installment due of the new modified loan (in particular, the borrower pays 40% of the

3

installment due of the new modified loan in the first year, 60% in the second year and 80% in the third year). In this case, the minimum interest margin charged over 3-month euribor, is set at 250 bps. This new modification program represents a concession for the Bank and is accounted for as TDR since it would not be considered by the Bank if the borrower was not in financial difficulty. The carrying amount of loans that were modified under this “fractional payment” modification program amounted to EUR · million and EUR 411.1 million at December 31, 2012 and 2011, respectively and were included in the total modified TDR mortgage loans referred to above. As at December 31, 2012, · borrowers had timely made payments under this modification program compared to · that have subsequently re-defaulted (December 31, 2011: 4,848 and 1,431, respectively).”

We also advise the Staff that the success rates of the fractional payment modification program for the period since inception (that is, mid-2011) through the year ended December 31, 2011, through the quarter ended March 31, 2012 and through the quarter ended June 30, 2012 are 74.2%, 56.7% and 47.6% respectively.

Finally, we advise the Staff that loans modified under the “fractional payment” modification program are included in the percentage of modified loans that were subsequently modified again as disclosed on page 142 of our Form 20-F for the year 2011.

Small Business Lending Unit, page 51

4.We note from your disclosure on page 53 that you classify all modified small business lending (SBL) loans as TDRs. Please address the following:

·Given that the modification terms and concessions offered are substantially similar to those offered on consumer and mortgage loans that are modified, please tell us why you consider all of your modified SBL loans to be TDRs and, conversely, why you do not consider all of your modified consumer and mortgage loans to be TDRs.

We advise the Staff that, as disclosed on page 142 of our Annual Report on Form 20-F for 2011, not all our modified SBLs are TDRs. Furthermore, as stated on page 52 of our Form 20-F for the year 2011, the SBL modification programs described mainly target borrowers with at least one account in delinquency of at least 90 days, that is, at least one of the customer’s accounts was non-accruing, and therefore part of the unrecognized interest income was waived upon modification, which constitutes a concession. Similar to our modified consumer and mortgage loans, the waiver of the unrecognized interest income constitutes a concession that would not be otherwise considered by the Group if the borrower was not in financial difficulty; therefore those loans were determined to be TDRs. Also, similar to our modified consumer and mortgage loans, modifications that do not involve a concession are determined not to be TDRs.

4

·Consider revising future filings to disclose the amount of SBL TDR loans you hold as of each period end presented, similar to your disclosures on page 49 of both non-TDR modified mortgage loans and mortgage loans classified as TDRs at December 31, 2011 and 2010. In your disclosure, please clarify whether these amounts include or exclude your restructuring-on-restructuring (R-O-R) SBL loans.

We advise the Staff that we will include the following in our next annual report on Form 20-F:

“As at December 31, 2012 and 2011, non-TDR SBL loans modified within 2012 amounted to EUR · million, compared to EUR 99.6 million modified within 2011. As at December 31, 2012 and 2011, modified SBL loans that were determined to be TDRs, including restructuring-on-restructuring products, amounted to EUR · million and EUR 504.3 million, respectively.”

5.Tell us whether loans modified under your R-O-R program are included in the percentage of modified loans which were subsequently modified again as disclosed on page 142. If they are not included, please revise your future filings to disclose that fact and explain why not.

We advise the Staff that loans modified under our R-O-R program are included in the percentage of modified loans which were subsequently modified again as disclosed on page 142 of our Form 20-F for the year 2011.

Corporate and Investment Banking, page 53

6.We note your disclosure here that, as a result of the ongoing economic crisis in Greece, you have adopted a more conservative approach to new commercial lending with a focus on larger corporate borrowers that you perceive to be lower-risk. Please tell us and revise this disclosure in future filings to expand the discussion of your new lending approach to describe the specific processes and procedures you follow when determining whether to lend to a commercial borrower. Please also address how you define “lower-risk” in this context, and identify the specific credit terms or other borrower characteristics you rely on to determine those that are lower-risk. Specifically address how these approaches consider the current macroeconomic pressures in Greece.

We advise the Staff that our lending process is in accordance with the Corporate Credit Policy that was first implemented in 2007, and revised in 2008 and 2011. The credit approval limits of the credit committees of the Bank have been reduced twice in the last two years (in 2011 and in 2012) to ensure increased senior management involvement in a larger number of credit evaluations. The Credit division participates in all credit committees and has veto power in all approval levels. Note that the loan approval process regarding corporate banking is described in detail on pages 128-129 of our Form 20-F for the year 2011.

5

In addition, we advise the Staff that during the life of a loan, there are frequent reviews of the exposure approved by the relevant credit committee, where any deterioration in the financial condition of the borrower is noted. These reviews can be annual, semi-annual or more frequent if necessary. Taking into account the borrower’s creditworthiness, the borrower can be classified into one of various categories, using criteria in accordance with the Corporate Credit Policy. Depending on the category of classification, the borrower is monitored semi-annually or more frequently and, in every review, an action plan that also takes into consideration the current macroeconomic conditions is approved by the relevant credit committee.

We further advise the Staff that we will amend our future disclosure, as follows:

“As a result of the ongoing economic crisis in Greece, the Bank has adopted a more conservative approach to new commercial lending, with a greater focus on larger corporate borrowers that it perceives to be lower-risk. Lower-risk customers are considered to be those who:

·Have a good credit history with the Bank, with other financial institutions as well as with suppliers, for example, no history of delinquency, no bounced cheques and no material events affecting customer’s credit worthiness;

·Have a size that allows for relative resistance to market conditions or are market leaders in their respective sector, for example, focus on large customers with turnover greater than EUR 50 million;

·Operate in sectors that are not as affected by the economic crisis and have good recovery/growth potential, for example renewable energy, oil and gas, shipping;

·Are explicitly supported by their shareholders/beneficial owners who contribute own funds when necessary;

·Have adequate collateral to offer, such as real estate, receivables and cash/cheques;

·Have satisfactory financial performance, given the current circumstances, EBITDA and profit levels , adequate cash flow, low leverage and total obligations, as well as net assets.

As a result, substantial deleveraging took place in relation to the SME balances during 2011 partially offset by increased financing to Large Corporate borrowers for energy sector projects. The Bank expects an additional deleverage of 3% in its lending portfolio during 2012 compared to 2011. At the same time, during the last two years, the collateral coverage has increased across the entire portfolio. Specifically, the collateral coverage for the top 30 largest exposures has increased from 15% to 27%.

6

The credit terms offered in new commercial loans depend on each borrower and include:

·Higher pricing to reflect current market conditions and higher cost of funding;

·Higher equity contribution (for example, up to 60% financing and the remaining has to be equity participation for new projects and investments);

·Higher collateral coverage.”

7.Your table on page 142 quantifies the amount of commercial loans modifications, and your disclosure on page 98 refers to the fact that a majority of your restructured corporate portfolio comes from your own modification programs versus from the mandatory restructuring products specified by Greek Law. However, your disclosure here does not appear to address the modification programs for your commercial loan portfolio. Please revise your future filings to more clearly discuss your commercial loan modification programs by including disclosure related to these programs similar to your loan modification and TDR disclosures included on pages 47 through 53 for your other loan portfolios.

We advise the Staff that our commercial loans portfolio segment consists of Small Business Loans (“SBLs”) and Corporate loans. The modification programs for our SBLs are disclosed on pages 52-53 of our Form 20-F for the year 2011. In our next Annual Report on Form 20-F, we will expand the discussion on corporate loans modification programs, as follows:

“Loan modification programs are offered to corporate borrowers that are considered viable (with a high potential of recovery) but who have been affected by the current market conditions. Prerequisites for modification are the borrower’s ability to provide additional collateral, ability of shareholders to infuse additional equity into the business and an increase in loan pricing. The types of modification are usually a mix of new amortization schedule tailored to current conditions and the borrower’s projected cash flow, the extension or not of tenor, depending on the borrower and its needs, as well as shift from short-term to long-term financing. Most of corporate modifications are determined to be TDRs given that these modifications were for the purpose of maximizing the collection of the amounts due and thus constitute a concession that would not otherwise be granted.

As at December 31, 2012 and 2011, total modified corporate loans that were determined to be TDRs, amounted to EUR · million and EUR 335.0 million, respectively. As at December 31, 2012 and 2011, non-TDR corporate loans modified within 2012 amounted to EUR · million compared to EUR 7.6 million modified within 2011.”

7

Exposure to European countries with significant economic and fiscal strains, page 73

8.We note your disclosure here related to exposure in the countries of Portugal, Ireland, Italy, Spain and Hungary. We refer you to our prior comment 8 from our letter dated August 25, 2011 and your response to that comment included in your correspondence dated September 30, 2011. Please address the following:

·In your prior response, you indicated that you would present your gross exposure by country segregated between sovereign, corporate institutions, financial institutions, retail, small business, etc., and you provided us with an example of your future filing disclosure. Please revise this disclosure in future filings to present your gross exposure in a manner similar to that shown in your prior response. Further, please clarify that the funded exposure amounts disclosed is your gross exposure.

We respectfully advise the Staff that we have amended the presentation of this disclosure in our Form 20-F compared to the one indicated in our prior response based on our reading of the guidance issued on January 6, 2012 by the Division of Corporation Finance of the SEC “CF Disclosure Guidance: Topic No. 4 — European Sovereign Debt Exposures”.  However, in response to your comment above we will amend our next Form 20-F to include the same disclosure as indicated in our prior response.

·Revise this disclosure in future filings to provide a narrative discussion of what constitutes your unfunded exposure as presented here. For example, in your prior response you indicated that these amounts are comprised of letters of credit and letters of guarantees. Please also consider presenting these unfunded exposure amounts segregated by type of exposure within your tabular presentation.

We advise the Staff that, as mentioned in our response to the bullet point above we will amend our next Annual Report on Form 20-F to include the same disclosure as indicated in our prior response.

·In your prior response, you indicated that you have purchased credit default swaps (CDS) to hedge bond exposure in Italy and Portugal and that you would discuss such CDSs in future filings. Please revise your future filings disclosure to include this discussion and to provide a more comprehensive discussion of your use of CDS relating to your exposure to these countries.

We advise the Staff that as at December 31, 2011, we did not hold any positions in bonds of financial institutions in Portugal and Italy and we had closed the respective credit default swaps that we had mentioned in our response dated September 30, 2011 to prior comment 8 from your letter dated August 25, 2011. For this reason such discussion was not included in our disclosure in our Form 20-F for the year 2011.

8

If such CDSs are held again as of December 31, 2012, we will amend our disclosure similar to the response previously submitted.

·We note that both your funded and unfunded exposure to Spain has increased significantly between 2010 and 2011. Specifically, your funded exposure increased from €5.4 million to €11.5 million and your unfunded exposure increased from €3.2 million to €15.7 million at December 31, 2010 and 2011, respectively. Please tell us and revise your future filings to discuss the drivers behind your increased exposure to Spain as well as any other significant changes in your exposures.

We advise the Staff that although our exposure to Spain has increased, as of December 31, 2011 it approximates 0.03% of the Group’s total assets and therefore, we do not believe there is any benefit to investors in discussing these movements as they result from our normal operations and are not material to the Group’s consolidated results of financial condition.

Strategy, page 108

9.We note your disclosure here related to your strategy for meeting the challenges of the current economic crisis. You state that your medium-term goal is to become a large regional financial institution through measured, organic expansion, focusing on high-growth countries in your region. We also note your statement on page 109 that you intent to continue to support your international activities in certain countries, such as Turkey, so as to enhance healthy earnings growth and the geographical diversification of your income. These statements appear to contradict with your statement on page 108 that, as part of your plan to limit dependence on official support and to increase capital, you intend to dispose of a minority stake in Finansbank as well as another SE European banking subsidiary and to sell non-core assets in foreign countries. Please revise future filings to more clearly explain your strategy with respect to both your current and any future planned international operations, growth, expansion, and dispositions.

We advise the Staff that we have sought to identify our medium-term goals and at the same time to identify the shorter term strategies and actions we have implemented in response to the exigencies of the financial crisis in Greece. The current short-term situation and actions add time and uncertainty to achievement of medium-term goals. However, for example, our previously announced intention to dispose of a minority stake in Finansbank is consistent with a strategy to focus on Turkey as a high-growth market. We will revise future filings to eliminate ambiguities and to reflect with more specificity the impact of then-current circumstances on our strategy.

9

E. Selected Statistical Data, page 112
Credit Quality and Risk Management, page 128
Mortgage Loans, page 130

10.We note your disclosure on pages 131 and 132 regarding loans you hold that have been subsidized or guaranteed by the Hellenic Republic. You state that the these loans have so far not been affected by the crisis affecting the government, but that further deterioration of the government’s finances may adversely affect its ability to pay the subsidies and honor the guarantees. Further, we note that these loans are not eligible for restructuring. Please revise future filings to disclose the amount of any of these loans that are past due or in default. Further revise your table on page 141 to separately quantify the amount of such loans that are past due and that are included in each the accrual and nonaccrual classifications in that table.

We advise the staff that we will amend our disclosure in our next annual report on Form 20-F, as follows:

“Subsidized mortgages

…

As at December 31, 2011 and 2012, the total amount of mortgage loans subsidized by the Hellenic Republic or OEK amounted to EUR 2,062 million and EUR · million, respectively. As at December 31, 2011 and 2012, loans that are over 90 days past due in principal or interest amount to EUR 84 million and EUR · million, respectively. Mortgage loans subsidized by the Hellenic Republic or OEK are not subject to restructuring.

Guaranteed mortgages

…

As of December 31, 2011 and 2012, the total amount of mortgage loans guaranteed by the Hellenic Republic amounted to EUR 1,411 million and EUR · million, respectively. As at December 31, 2011 and 2012, loans that are over 90 days past due in principal or interest amount to EUR 49 million and EUR · million, respectively. Mortgage loans guaranteed by the Hellenic Republic are not subject to restructuring.”

We also advise the Staff that in future filings we will disclose the following table below the existing table on page 141:

10

Subsidized and guaranteed domestic mortgage loans

	Year ended December 31,
	2011	2012
Domestic loans:
Accruing loans:
Past due 90 days—180 days	63,464	·
Past due 180 days—365 days	—	·
Total accruing loans	63,464	·
Non-accruing loans:
Past due 0 days—90 days	—	·
Past due 90 days—180 days	—	·
Past due 180 days—365 days	22,721	·
Past due 365 days or more	46,046	·
Total non-accruing loans	68,767	·

Non-accruing Loans, Allowance for Loan Losses, and Loan Loss Experience, page 140
Non-Accruing, Past due and Restructured Loans, page 140

11.We note from your tabular disclosure on page 141 that total non-accruing loans increased 78.5% between 2010 and 2011, with a majority of the increase due to domestic non-accruing loans. Please revise future filings to thoroughly explain fluctuations in your non-accrual loans balances from period to period, including the reductions and additions to the balance. As part of such disclosure, clearly separately quantify the amount of loans removed from nonaccrual status and returned to accrual status because they were modified versus those loans that became current without modification.

We advise the Staff that the increase in our non-accrual loans is explained on page 171 and 172 of our Annual Report on Form 20-F for the year 2011:

Domestic non-accruing loans, page 171-172:

“The increase in non-accruing domestic loans as at December 31, 2011 of EUR 4,375.5 million is due to the following two factors: (a) the deterioration of the quality of our portfolio and (b) the effect from classification in 2011 of loans modified in a trouble debt restructuring as non-accruing, amounting to EUR 2,023.9 million as at December 31, 2011.”

Foreign non-accruing loans, page 172:

“Foreign non-accruing loans as at December 31, 2011, amounted to EUR 2,743.3 million increased by EUR 666.2 million or 32.1% compared to December 31, 2010. The increase is mainly attributed to the effect from classification in 2011 of loans modified in a trouble debt restructuring as non-accruing amounting to EUR 487.2 million as at December 31, 2011.”

11

We also advise the Staff that non-accruing loan balances may decrease as a result of write-offs, partial repayment of the past-due amount while the loan remains non-accruing, the return to accruing status without modification due to improved performance or post modification when a specified minimum number of payments are made by the borrower (ranging from 4 to 12 depending on the loan product), or the sale of non-accruing loans. The roll-forward of impaired loans disclosed on page F-58 of our Form 20-F quantifies the reduction of impaired loans due to all the above reasons. The main differences between non-accruing loans and impaired loans on the domestic portfolio are summarized below:

	2011
	(EUR in thousands)
Total domestic non-accruing loans	8,723,769
Loans to the Greek state	5,160,819
Loans eligible to PSI	1,779,022
Loans to public sector entities	394,228
Loans still accruing but impaired	493,627
Total domestic impaired loans	16,551,465

We advise the Staff that we will add the following disclosure under our tabular disclosure on page 141 in our next annual report on Form 20-F as follows:

“For the year ended December 31, 2011 and 2012, the amount of loans that have been modified in the past and were removed from nonaccrual status and returned to accrual status amounted to EUR · million and EUR · million, respectively.

For the year ended December 31, 2011 and 2012, the amount of loans that were removed from nonaccrual status and returned to accrual status because they became current without being modified in the past amounted to EUR · and EUR ·, respectively.”

Modifications and TDRs, page 141

12.We note your disclosure on page 142 of TDRs granted during the last 12 months, the percentage of TDRs granted in the last 12 months that subsequently became past due +30 days, and the percentage of modified loans which were subsequently modified again during the year ended December 31, 2011. Please address the following:

·We note from your tabular disclosure on page 142 that modified loans increased 19.5% and TDR loans increased 36.1% between 2010 and 2011. Please revise future filings to discuss the reason(s) for fluctuations in your modified loans and TDR loans balances from period to period.

12

We respectfully advise the Staff that the percentage of modified and TDR loans increased by 24.2% and 56.5%, respectively, between 2010 and 2011. In our next Annual Report on Form 20-F we will add the following disclosure on page 142 to discuss the reasons for the fluctuations in our modified and TDR loans:

“Modified and TDR loans increased by 24.2% and 56.5%, respectively, between 2010 and 2011 due to (a) the fact that modification programs began after March 2010 and therefore data related to year ended December 31, 2010 refers to a shorter period of production compared to 2011, and (b) efforts were enhanced through 2011 to continue the modification programs and the launch of new programs such as “the Fractional payment” modification program.”

·We note that the re-default rates on TDRs and other loans modified during the last 12 months was substantial for both your consumer and commercial loan portfolios. In future filings please more clearly disclose what steps you take and what criteria you consider in determining which loans to modify and whether a given loan will subsequently re-default.

We advise the Staff that we will add the following disclosure in our next Annual Report on Form 20-F:

“The Group has loan modification policies for all loans portfolios, which describe the eligible population based on certain criteria considered in determining which loans to modify and whether a given loan will subsequently re-default.  The main criteria considered when modifying retail loans remain the number of days past due and the ability of the borrower to provide additional collateral. For corporate loans, in addition to the above we also consider the payment history of the borrower and the reasons for the deterioration of the borrower’s financial condition. Especially for corporate loans modification programs, it is important to note that, when modifying a loan, the new terms reflect the borrower’s financial condition, its ability to make future payments, the current and projected market conditions, as well as the factors discussed in Item 4.B. “Business Overview-Banking Activities in Greece-Corporate and Investment Banking”. All modification programs are subject to credit committee approval and follow the lending approach described in the Credit Policies.”

·Clearly explain how and the extent to which managing your re-default rates is a consideration in the design and implementation of your various loan modification programs.

We advise the Staff that we continuously evaluate our modification programs, taking into consideration their success and the changing economic condition in the country.

13

The trends of re-default are closely monitored and analyzed in order to identify the drivers for the re-defaults. In certain cases, monitoring and assessment of the payment history of modified loans led us to adjust our modification policy, for example, by introducing, for all new modifications of SBLs the mandatory receipt of property collateral (wherever available), as well as a further reduction of the installment amount, either through the reduction of the interest rate or to the extension of the repayment period by introducing duration up to 25 years.

In addition, the significant re-default rate observed was the reason we designed the fractional payment modification program offered to mortgage or SBL customers with loans that have been modified and have re-defaulted. These fractional payment modification programs aim to further decrease the amount of the monthly installment to a viable level, after taking into consideration the decreasing income of the borrower.

In 2012, the mortgage lending division re-evaluated the 3-year fractional payment modification program and further reduced the monthly installment payable for the first three years, based on the delinquency of the loans modified.

·Discuss in detail how you consider TDRs and other loan modification programs in determining the amount of your allowance.

We advise the Staff that TDRs are separately assessed within each portfolio for the purposes of allowance for loan loss calculation. Allowance for loan loss is calculated based on a present value of expected future cash flows methodology considering all available evidence at the time of the assessment. Typically, allowance for loan loss for TDR Consumer loans are calculated using the average of yearly default frequencies of those specific products, while allowance for loan loss for TDR corporate loans is usually calculated on an individual basis given that the restructuring is an impairment trigger for individual assessment. For TDR corporate borrowers that do not satisfy the quantified criterion for individual assessment (that is, the individually non—significant borrowers), allowance for loan loss is calculated through our coefficient analysis using the probability of default corresponding to their internal credit rating.

Modified loans that are not TDRs are assessed similarly to non-modified loans when determining the amount of the allowance for loan loss through homogenous assessment based on credit risk parameters depending on the allowance for loan loss methodology.

·Identify how you consider the trends in your re-default rate in determining the appropriate level of the allowance.

We advise the Staff that trends of re-default are generally considered when calculating the appropriate level of the allowance for loan loss by adjusting the probabilities of default relating to the TDR loan.

14

·Discuss whether a re-default triggers a higher allowance for a given loan or whether the likelihood for re-default is built into your allowance at the time of the first loan modification.

We advise the Staff that TDR loans will have been generally previously identified as impaired loans prior to the modification. Upon modification, the estimate of the allowance for loan loss of a TDR loan generally considers the re-default probability of those loans. That is, the likelihood of re-default is built into our allowance at the time of the first loan modification. However, upon a loan’s re-default, the allowance is increased since it is now a defaulted loan and the likelihood of default is increased to 100%.

·You state that the percentage of modified loans which were subsequently modified again was 7.2% for the year ended December 31, 2011. Please revise future filings to clarify whether these loans were included in the percentage of modified loans granted in the last 12 months that subsequently became past due +30 days, or whether these loans are in addition to those other percentages.

We advise the Staff that in future filings we will revise our disclosure, as follows:

“Modified and TDR loans are separately monitored by dedicated teams. Special reports are produced for modified and TDR loans and in case of non-payment, they are subject to more intensive collection actions. As at December 31, 2012, the percentage of TDR loans granted during the last 12 months that became past due (+30 days past due) was ·% for consumer loan portfolio and ·% for commercial loan portfolio, and ·% and ·%, respectively for modified loans.

New modification programs for previously modified loans were first launched during the third quarter of 2011. As of December 31, 2011 and 2012, the percentage of modified loans which were subsequently modified again was 7.2% and ·%, respectively. These loans are included in the percentage of TDR granted in the last 12 months that became past due (+30 days past due) after modification only if their last modification occurred during the last 12 months and they have become past due (+30 days past due) subsequent to re-modification.”

15

Item 5. Operating and Financial Review and Prospectus, page 150
Other-Than-Temporary-Impairment of other available-for-sale and held-to-maturity financial assets, page 168

13.Please address the following regarding the various debt securities that have been in an unrealized loss position for greater than 12 months:

·Tell us and revise this section in future filings to more clearly explain how you assess the credit quality and collectability of these debt instruments, to identify the credit quality indicators you use, and to discuss the indicators assigned to those companies.

·Clearly identify the positive and negative evidence you considered in such a manner that gives proper weight to both.

·You discuss your conclusions of whether to classify your holdings in other Greek banks as other-than-temporarily-impaired. Tell us and revise your future filings to quantify the amount of such debt securities you held as of the balance sheet dates presented.

We advise the Staff that in our next Annual Report on Form 20-F we will add the following disclosure:

“For the purposes of assessing the credit quality and collectability of the debt instruments, the Group segregates those securities classified as available-for-sale and held-to-maturity first by issuer type between Sovereign, Financial Institutions and other corporate entities. Each issuer type is then further disaggregated by region to Eurozone, European Union (non-Eurozone), Outside the European Union and Emerging Market (countries that have been labeled by the International Monetary Fund as “emerging economies”). Corporate entities are further disaggregated into listed and non-listed. Evidence considered is dependent on the category of the debt. With respect to our AFS debt securities issued by other governments and public entities, in an unrealized loss position for 12 months or more, the fair value as at December 31, 2011 amounted to EUR 60.0 million with unrealized loss of EUR 16.7 million. EUR 51.6 million of these securities with unrealized loss EUR 15.8 million related to the Republic of Cyprus. In deciding that the securities are not other-than-temporary-impaired we took into consideration the following indicators:

1.The general financial situation of the sovereign, which in the Cyprus case was negative since the financial situation in Cyprus is deteriorating due to problems in its banking system that affects the sovereign. However, we considered the fact that this was more of a banking crisis rather than a structural problem related to the country was a mitigating factor with respect to the credit risk of the sovereign.

2.Ability to access the bond markets, which in the Cyprus case was a positive indicator since the sovereign had the ability to access the bond market for funding.

16

3.Its ability to access the European Support Mechanism (the “ESM”), which in the Cyprus case was positive since as a member of the Eurozone, Cyprus has the ability to access the ESM resources through the implementation of an adjustment program thus increasing its liquidity and its ability to make payments on its debts in the future.

4.Denomination of the debt, which in the Cyprus case was positive since the debt is denominated in Euro and as such, there is no FX risk since its tax receipts are also denominated in Euro.

We concluded that the most important factors in our analysis was that Cyprus has the ability to access the ESM mechanism and thus can avoid a liquidity crisis and as such the debt was not other-than-temporarily-impaired as at December 31, 2011.

With respect to debt securities issued by companies incorporated in Greece the loss of EUR 232.6 million includes EUR 182.6 million relating to debt securities issued by the three viable financial institutions in Greece, other than the Bank (that is, Alpha Bank, Eurobank and Piraeus Bank) as judged by the Troika, and EUR 47.8 million relating to OTE, the largest telecom provider in Greece, which is owned by Deutsche Telecom S.A.

Indicators which we used for financial institutions incorporated in Greece include:

1.Is the financial institution judged to be viable and meets all the other criteria set for the recapitalization by the Hellenic Financial Stability Fund?  This was a positive indicator since all three have been judged to be viable and will be recapitalized by the HFSF by April 20, 2012.

2.Does the institution have the ability to access the interbank market? This was a negative indicator since the financial institutions mentioned above have been locked out of the international interbank market and the interbank market in Greece was operating at a fraction of its volume due to the crisis.

3.Does the institution have the ability to access the Bank of Greece’s (the “BoG”) Emergency Liquidity Assistance (“ELA”) funding program? This is a positive indicator since all the financial institutions have access and made use of the BoG’s ELA funding to ward off liquidity problems.

We determined that the most important factors are that the financial institutions have been judged to be viable and eligible for recapitalization by the HFSF and thus will be recapitalized, and their ability to access the BoG’s ELA funding to ward off any liquidity crisis until their recapitalization. As such we concluded that an other-than-temporary-impairment had not occurred as at December 31, 2011.

17

OTE is the largest telecom provider in Greece and is listed on the Athex and based on its market capitalization is grouped in FTSE/ATHEX 20 Index (largest 20 Corporations in Greece measured by market capitalization). Some of the indicators we examine include:

1.Profit before tax, which for OTE shows a positive trend as profit before tax increased by 15% in 2011.

2.Cash flow generation. Although the cash flow was negative between 2010 to 2011 it was mostly attributable to the repayment of debt in 2011.

3.Ability of a parent corporation to support the subsidiary. For OTE this is positive since Deutsche Telecom is the biggest Telecom operator in Germany and can provide liquidity support to OTE in order not to impair its investment.

4.Economic environment. For OTE this was negative since its primary operating environment has been in recession for the last 5 years.

5.Other items- This was positive for OTE since they proceeded with the repayment of a EUR 1,370.7 million bond that came due in February 2011 as well as the fact that they have purchased their bonds in the secondary market at a discount and retired them.

Based on the fact that OTE is profitable before tax and has repaid a major bond maturity in February 2011 and is also active in the secondary market buying its debt at a discount and retiring it, management concluded that the impairment was not other then temporary in nature.

With respect to debt securities issued by companies incorporated outside Greece the loss of EUR 77.0 million relates to debt securities issued mainly by major European and American banks and insurance companies with an unrealized loss of EUR 66.7 million and EUR 9.8 million, respectively. We consider that these banks will continue to pay in accordance with the contractual terms and as such an other-than-temporary-impairment was not considered necessary as at December 31, 2011.”

Operating Results, page 170
Net Interest Income before provision for loan losses, page 170

14.Please revise future filings to address the following in order to provide a more detailed discussion of the changes in your net interest margin:

·More clearly discuss the reasons for the increase in interest expense on deposits, as well as your expectations regarding trends in this area.

18

·Discuss and quantify your reliance on wholesale deposits and brokered deposits during the periods presented as well as the extent to which you expect such trends to continue.

·Clearly quantify the impact of any changes in the mix of your funding sources on your net interest margin, as certain sources become more or less available to you.

·Further, please discuss how you expect the significant increases in your nonaccrual balances will affect your net interest margin in future periods.

We advise the Staff that we will add the following disclosure in our next Annual Report on Form 20-F:

“The increase in interest expense on deposits during 2011 relates to (i) the significant higher rates offered on time deposits in Greece in various customer segments amidst increasing market competition following the adverse economic conditions and the liquidity shortage in the domestic market, (ii) the higher rates paid for the ELA facility introduced in the 2nd half of 2011 in comparison to cheaper funding from ECB. This trend of increased deposit rates is expected to continue in the following year however at a slower pace.

As at December 31, 2011 wholesale deposits (defined as interbank deposits, which include ECB and other central banks) represented approximately 36% of the total deposit pool compared to 28% as December 31, 2010, while the Group has no brokerage deposits.

The Group’s net interest margin is dependent on disciplined deposit pricing and asset repricing efforts. In domestic market retail deposits and more specifically savings deposits which are considered as a cheap funding source become less available for the Bank given the prevailing adverse economic conditions in the domestic market. This trend negatively affected net interest margin on the expense side in two ways: (i) increased the reliance on funding from the ELA facility (EUR 9.1 billion as of December 31, 2011), which has a higher cost almost similar to a time deposit and (ii) increased the contribution of costly time deposits in total deposit mix. On the other hand increased loans spreads charged to retail and corporate lending customers reflecting the scarcity of funds and heightened credit risk compensates for the margin compression on the liability side.  Furthermore, our net interest margin is also dependent on the increased weight the foreign activities have over the reducing domestic balance sheet.

In addition, the increase of nonaccrual balances had a negative impact on our net interest margin during 2011 and this trend is expected to continue in future periods, especially in Greece, while the economy is under recessionary pressures.”

A similar disclosure will be included, in our next Annual Report on Form 20-F to address the fluctuation in our net interest margin in 2012 and our expectations regarding trends in this area, including our interest expense on deposits.

Liquidity and Capital Resources, page 188
Liquidity Management, page 188

15.Revise this section to discuss in greater detail the other aspects of the PSI exchange, the recapitalization, and other related terms imposed that you believe will affect your liquidity situation. Specifically tell us and revise to discuss the terms of such agreements that require reduction in your holdings of Greek Government Bonds and/or treasury bills.

We advise the Staff that under the terms of the PSI exchange, our holdings of certain specified outstanding Greek government bonds (“old GGBs”) were exchanged for bonds issued by the European Financial Stability Fund (the “EFSF”) (with face value of 15% of the face value of the old GGBs) (“EFSF

19

Bonds”) and new GGBs (with face value of 31.5% of the face value of the old GGBs). The EFSF Bonds and the new GGBs are tradable, and actually by June 30, 2012 we have sold at approximately the face value most of the EFSF Bonds received in the PSI exchange. In addition, both the EFSF bonds and the new GGBs are eligible for repo financing from the European Central Bank (the “ECB”), in order to raise liquidity.

We also advise the Staff that recapitalization of the Greek banks has not been completed yet and its actual terms have not been finalized. However, the funds necessary for the recapitalization will be provided to the Bank in the form of EFSF bonds, which although the Bank will not be allowed to sell, we will be allowed to use them as collateral for repo financing from the ECB. Also note that the Bank received in May 2012 EFSF bonds of EUR 7.4 billion as pre-funding in the context of the contemplated recapitalization, and these EFSF bonds are used as collateral for repo financing from the ECB.

As of this date, we are not aware of any other related terms that have been or are expected to be imposed on us that would affect our liquidity. Specifically, we have not entered into any agreements that require us to reduce our holdings of GGBs or treasury bills.

Capital Management, page 191

16.Please revise to more clearly quantify the decreases in your deposits exclusive of your interbank funding, and in particular excluding amounts from entities such as the ECB and other central banks. Discuss the comparative funding rates between retail customers, brokered deposits, interbank deposits, and central bank funding. Quantify and identify the nature of collateral available and used for central bank funding.

We advise the Staff that we will revise the relevant disclosure in our next Annual Report on Form 20-F as follows:

“Deposits decreased from EUR 93.9 billion on December 31, 2010 to EUR 91.6 billion on December 31, 2011 and decreased to EUR · billion on December 31, 2012. Excluding interbank funding (which includes funding from the ECB and central banks), deposits decreased from EUR 67.8 billion on December 31, 2010 to EUR 58.9 billion on December 31, 2011 and decreased to EUR · on December 31, 2012. Interbank funding, including funding from the ECB and central banks, increased from EUR 26.1 billion on December 31, 2010 to EUR 32.7 billion on December 31, 2011 and to EUR · billion on December 31, 2012.

As discussed in Item 5.A. “Operating Results-Results of operations for the year ended December 31, 2011 compared with the year ended December 31, 2010-Net Interest Income before provision for loan losses”, the decrease in customer deposits during 2011 resulted in increased funding cost due to the reliance in Eurosystem funding through ELA and the higher cost of time deposits (see Item 4.E. “Selected Statistical Data-Liabilities-Deposits” for an analysis of our funding rates per type of deposit).

As at December 31, 2012, collateral with nominal value of EUR · billion (December 31, 2011: EUR 48.1 billion) was used for central bank funding. This relates to debt securities, covered bonds, securitizations, customer loans and bonds issued from our participation in the Hellenic Republic Bank Support Plan. As at December 31, 2012 the nominal value of the available collateral for central bank funding was EUR · billion (December 31, 2011: EUR 11.0 billion) relating to debt securities and loans.”

20

Item 6. Directors, Senior Management and Employees
A. Board of Directors and Senior Management, page 212
Greek government influence, page 216

17.We note from disclosures in Form 6-K filings that the Chairman of your Board of Directors, your Chief Executive Officer, and your Chief Financial Officer all resigned in June 2012. We also note that recent information in the press indicates that top officers are typically replaced upon a change in the Greek government. It appears that the ability for such influence predates and significantly exceeds the government influence discussed in this section. As such, please address the following:

·Revise to clearly identify the Board members that exercise Greek government influence as discussed in this section.

We advise the Staff that Mr. Alexandros N. Makridis, the Greek government representative in the Bank’s Board of Directors in the context of Pillar I of the Hellenic Republic Bank Support Plan, is identified in the table on page 216 of our Form 20-F. The powers of this representative are clearly described in Item 6.A, section “Greek government influence” on page 216 of our Form 20-F, as well in Item 3.D “Risk Factors” on page 37, Item 4.B, section “The Hellenic Republic Bank Support Plan” on page 89 and Item 10.J “Relationship with the Hellenic Republic” on page 268. Therefore, we do not consider necessary to revise our Form 20-F in this respect.

Furthermore, in the context of the recapitalization of Greek banks, the Hellenic Financial Stability Fund (the “HFSF”) has appointed, on June 11, 2012 (that is after the filing of our Form 20-F for the year 2011), Mr. Charalambos Makkas as its representative in the Bank’s Board of Directors, pursuant to Law 3864/2010, art 6.10. The possibility of the Greek Government exercising a greater level of control following the recapitalization by the HFSF is disclosed in Item 6.A., section “Greek government influence” on page 217 of our Form 20-F, and the rights of the HFSF representative are disclosed in Item 4.B, section “Recapitalization Framework” on pages 80-81. The HFSF representative will be identified in our next Annual Report on Form 20-F similarly to the Greek government representative, and his powers will also be described. Furthermore, the appointment of Mr. Makkas as the HFSF representative in the Bank’s Board of Directors was included in our Form 6-K filing dated June 13, 2012 and has also been disclosed on the Bank’s website.

Pursuant to the above mentioned laws, both the Hellenic Republic and the HFSF have certain powers over credit institutions receiving financial support, among which includes a veto power of their representatives on the Board of Directors’ decisions. This power, however, does not extend to the decisions of the Board regarding the nomination of a Board member, and this power is not provided for anywhere in the law.

21

According to the above mentioned laws, such veto power is limited and specifically granted on:

·strategic decisions or decisions resulting in a significant change of legal or financial position of the Bank (for example mergers, acquisitions) and for which shareholders’ approval is required, and

·decisions relating to the dividend policy and the compensation of the Chairman, the Managing Director and the other members of the Board of Directors of the participating bank, as well as its General Directors and their deputies.

Apart from their above representatives and the rights of the Hellenic Republic as a shareholder, both the Hellenic Republic and the HFSF do not have other powers or control over the appointment of any other member of the Bank’s Board.

We also advise the Staff that the Hellenic Republic has appointed, since 2009, a representative in each of the eight Greek banks that have participated in the Hellenic Republic Bank Support Plan, and the HFSF has appointed a representative in each of the four major Greek banks that will be recapitalized through HFSF, pursuant to Law 3864/2010.

·Please revise future filings to disclose the screening process your Board of Directors uses to evaluate candidates for appointment to top executive officer positions.

We advise the Staff that we will revise the disclosure in our next Annual Report on Form 20-F, as follows:

“Directors Nomination

The nomination procedure and the qualification criteria for Directors are specifically provided for in the Bank’s internal corporate governance rules and especially: a) the Bank’s Articles of Association, b) the NBG Group Corporate Governance Code, c) the Board’s Corporate Governance and Nomination Committee Charter and d) the Directors’ Nomination Policy. Each nominee fulfills such criteria that ensure the appropriate governance and guidance of the Bank’s strategy in relation to economic, business and policy issues.

According to article 9.1.b of the Bank’s Articles of Association, the General Meeting of Shareholders is the sole corporate body vested with the authority to decide on the nomination of members of the Board of Directors. Furthermore, according to the applicable Greek legal framework (Law 2190/1920), which applies to all Greek private companies (that is, companies not owned by the state), including the Bank, since it is a private company, and the Bank’s Corporate Governance Code, any shareholder may propose a candidate for Board of Directors membership at the General Meeting of Shareholders and this right (that is to propose a candidate at the General Meeting) is

22

also given to the Board of Directors. To this end, the Board of Directors, assisted by the Board of Directors’ independent Corporate Governance and Nomination Committee, proposes to the General Meeting candidates for Board of Directors membership on the basis of the Director’s Nomination Policy that requires them to meet specific qualifying criteria (“fit and proper”), according to the Bank’s Corporate Governance Code. It is not required for the Board to submit a separate proposal for the positions of the Chairman of the Board of Directors and the Chief Executive Officer.

Following the election of the new members of the Board of Directors by the General Meeting of Shareholders, the new Board of Directors in its inaugurating session elects by absolute majority, from its members, the Chairman of the Board of Directors and the Chief Executive Officer who manages the Bank. Moreover, the Board of Directors may elect Vice Chairman and Deputy Chief Executive Officers. The Chairman and the Chief Executive Officer may be the same person, and the same applies for the Vice Chairmen and the Deputy Chief Executive Officers.

In discharging their nomination responsibilities and proposals to the General Meeting of Shareholders or in appointing new members in replacement of retired members, the Board of Directors proposes candidates to be appointed as Directors or replace members of the Board of Directors that have submitted their resignation, who possess a particular profile, in compliance with the NBG Directors’ Nomination Policy and whose nomination ensures that the Board of Directors as a collective body presents the following profile:

·has in depth knowledge of the financial industry, counting amongst its members individuals that are or have been active in leadership positions in financial institutions;

·possesses significant business and professional experience and is socially distinguished, including, among its members, individuals that are serving or have served as Chairmen, Chief Executive Officers or senior managers of large organizations and have built a reputation that demonstrates the ability to form judgments over important and sensitive matters upon which the Board of Directors is called to decide;

·has a full understanding of the Bank’s client base structure and dynamics, and of the main geographic markets in which the Bank is currently active;

·has considerable international experience and is able to contribute to NBG’s aspirations in the geographical region in which the Bank is active;

23

·has the financial expertise required to provide effective oversight of a diversified financial services Group that operates on an international scale; and

·ensures, as far as it is possible, an adequate representation of the two genders.

The NBG Group Corporate Governance Code, as well as the Directors’ Nomination Policy provide for specific qualifying criteria (“fit and proper”) for the nomination of candidates; professional capacities that are incompatible with the position of Board Director at NBG; criteria for non-executive directors’ independence; and finally cases of conflicts of interest and maximum number of allowed other appointments/ external activities of a director. All the above aim to ensure and guarantee the best composition for the Board of NBG.

The persons elected in 2011 as new members of Board of Directors and as top executive officers of the Bank meet fully the qualifications and high standards set by the Bank in its internal corporate governance framework, have substantial business and professional experience and have built a reputation that demonstrates their ability to make the important and sensitive business decisions that the Board of Directors and top executive officers are called upon to make. All new members of the Board of Directors and top executive officers were already serving the Bank (or had been in the past) through various levels of seniority, having in this way a full understanding of the structure and dynamics of the NBG Group and of the principal markets in which the Group is currently active. In particular, Mr. G. Zanias (Chairman of the Board) had also been elected in the past (January 2010) as a member of the Bank’s Board of Directors, Mr. Alexandros Tourkolias (CEO) was already a member of the Board and Deputy CEO since November 2010, as well as a member of the Board’s Executive Committee since 2006 (previously and for 15 years holding several managerial positions in NBG), while Mr. Petros Christodoulou (sole Deputy CEO) was recently General Manager of International Activities and also NBG Group Treasurer since 1998. All the above are eminent members of the Greek business community and society (also refer to their curricula vitae below).”

·Address the degree to which your Board relies on outside parties in making their appointment decisions, and the extent to which those outside parties including the Greek government are able to influence the appointment of new executive officers through the government’s representatives on the Board or outside that process. Clearly explain how such influence is exercised.

We advise the Staff that, other than by the rights as a shareholder (as described in our response to bullet point 2 above) and the two representatives of the Greek government and the HFSF (as described in our response to bullet point 1 above), our Board of Directors does not rely to outside parties in making their appointment decisions, and the Greek government, or any other third party is not able to influence the appointment of new Board of Directors members.

24

In particular, as described above, the powers of the Greek government and the HFSF (and their representatives) do not include veto power or control over the appointment of any other member of the Bank’s Board of Directors or the election of new executive officers.

·If outside parties or the Greek government are able to influence executive officer appointments, specifically disclose whether the executive officer selections made by the outside parties are subject to Board approval. Clearly identify the extent to which the Board retains the ability disapprove appointments made by outside parties, both in form and in function.

We advise the Staff that, as described in our response to bullet point 3 above, no outside parties are able to influence executive officers appointments.

Item 11. Quantitative and Qualitative Disclosures About Market Risk
Value-at-Risk (VaR), page 270

18.We note your disclosure that you have established a framework of Value-at-Risk (VaR) limits in order to control and manage more efficiently the risks to which you are exposed, and that you perform back-testing on a daily basis in order to verify the predictive power of you VaR model. Please respond to the following and expand your disclosure in future filings as appropriate:

·Tell us how all of the different VaR models used are aggregated to arrive at your total consolidated trading and AFS VaR as well as the individual market risk categories disclosed. For example, clarify whether you simply aggregate the outputs from the different models or whether adjustments are made, and if so, how the adjustments are determined.

The VaR model used by the Bank is the same for all the consolidated subsidiaries’ portfolios and for all the individual market risk categories disclosed (Variance-Covariance methodology) and the aggregation is based on the well known RiskMetrics approach.

·Clarify whether your stress VaR scenarios, on a consolidated basis, use all of the base VaR models with more stressed assumptions or if you use different models than your unstressed VaR scenarios. To the extent that different models are used to capture that risk, please tell us the drivers as to why.

The calculation of Stressed VaR is based on exactly the same methodology used for calculating the non-stressed VaR, but applying more stressed assumptions as described

25

below.

In particular, the Stressed VaR is a measure intended to replicate a value-at-risk calculation that would be generated from the bank’s current portfolio, if the relevant market factors were experiencing a period of stress.

It is based on:

·the 10-day, 99th percentile;

·one-tailed confidence interval value-at-risk measure of the current portfolio;

·model inputs calibrated to historical data from a continuous 12-month period of significant financial stress relevant to the bank’s portfolio.

·We note your disclosure on page 271 related to changes in VaR in 2011 due to charges in hedges and interest rate exposure. Clarify whether you made any changes to your VaR model or assumptions during 2011. Discuss the process and validation procedures in place prior to implementing significant model and assumption changes. For example, discuss the approval process required, back-testing procedures performed, and periods of parallel model runs before implementation.

We advise the Staff that we made no changes to our VaR model during 2011. The changes in the VaR estimates were only due to changes in positions and/or relevant market risk factors.

All VaR model components, including input, processing and reporting have been subject to specific validation conducted by external advisors and internal independent units and has been approved by our regulator, the Bank of Greece.

All VaR model components are validated, on a yearly basis or whenever there is a material change. The validation framework includes the following steps:

1)Evaluation of conceptual soundness, which concerns assessing the quality of the model design and construction;

2)Ongoing monitoring in order to evaluate whether changes in products, exposures or market conditions necessitate model redevelopment;

3)Outcome analysis or back testing, which consists comparing the actual profit and loss with the model forecast;

·Identify whether the VaR models discussed here are the same models used for regulatory purposes, and if not, discuss the nature of the differences between the models.

26

The internal VaR model disclosed in this section is different from the Regulatory VaR model. The differences between the two VaR models are the following:

1)The internal VaR model is based on the variance-covariance methodology and is calculated using a weighted covariance matrix while the Regulatory VaR model, although it is also based on the variance covariance methodology, it uses an un-weighted covariance matrix;

2)The internal VaR includes both the Trading and the Available for Sale portfolios, while the Regulatory VaR includes only the Trading portfolio;

3)The internal VaR includes the general and specific risk of equities, while the Regulatory VaR includes only the general risk of equities.

·Given that your VaR models are evolving over time, tell us how you consider when disclosure is required under Item 305(a)(1)(iii)(4) of Regulation S-K regarding model, assumptions and parameter changes.

We advise the Staff that our VaR model has not changed during 2010 and 2011, therefore we consider that the disclosures required by Item 305(a)(1)(iii)(C)(4) were not applicable.

Limitations of our VaR model, page 272

19.We note your response to comment two of our letter dated April 9, 2012 regarding VaR as well your disclosure here. Please revise your disclosure in the last bullet of this section that your model excludes the credit risk of financial instruments to address the following:

·Confirm in your future disclosure that all of the fair value inputs into your VaR model include any inherent credit risk. If not, explain where such credit risk is excluded and why.

We advise the Staff that, in 2012 we changed our VaR model to incorporate the credit risk. As a result in our next Annual Report on Form 20-F we will revise our VaR disclosures for 2011 based on the new VaR model, in accordance with Item 305(a)(1)(iii)(C)(4). Also, in our next Annual Report on Form 20-F we will add the following disclosure:

“All fair value inputs into our VaR model include any inherent credit risk.”

·Tell us and revise to disclose how you determined that the historic returns of the German benchmark bonds were the appropriate substitute for the various financial instruments in your portfolio covered by your VaR models.

We advise the Staff that the Internal VaR disclosed, as explained in the section “Limitations of our VaR model”, measured the interest rate risk and did not incorporate

27

the credit spread of financial instruments. This is the reason that the German benchmark bond yields were used as a substitute for the calculation of the volatility of the various financial instruments.

As mentioned in our response to the bullet point above, starting from the second quarter of 2012 we will change our VaR model to incorporate the credit risk.

Item 18. Financial Statements
Notes to Consolidated Financial Statements
Note 11: Investment Securities, page F-39
Greek bonds eligible for the PSI, page F-47

20.You disclose on page F-48 that there was not an active market for Greek government bonds (GGBs) at December 31, 2011 and you estimated the fair value based on the contractual cash flows of the new GGBs, a discount rate of 12%, and market information. You further disclose on page F-108 that you classified these GGBs as Level 3 as of December 31, 2011. Please address the following:

·Tell us how you considered the guidance in paragraphs ASC 820-10-35-51A and 51B prior to the adoption of ASU 2011-04 in determining that there was not an active market for these GGBs.

We advise the Staff that we considered the guidance in ASC 820-10-35-51A and 51B in determining that the volume of activity for the PSI eligible bonds had decreased significantly. More specifically, we relied on external/market information for ASC 820-10-35-51A (a) and (f) as follows:

1.HDAT (the Bank of Greece’s platform for the electronic trading of Greek government bonds): the HDAT records available on the Bank of Greece website show that there were EUR 2 million in trades in GGBs in the entire month of December 2011 (nil trades for PSI eligible bonds held by the Group) which indicates the market was inactive per ASC 820-10-35-51A (a).

For the Staff’s reference we are including volume information in HDAT for the last three years to demonstrate the significant reductions in volumes by December 31, 2011:

	Trading volumes
	EUR millions
Period	2011	2010	2009
Q4	3	3,136	123,113
Q3	177	4,102	100,744
Q2	2,467	13,543	60,302
Q1	2,528	74,558	42,203
Total	5,175	95,339	326,362

28

2.Other exchanges: we also sourced information regarding transaction volumes from the Frankfurt exchange using the Bloomberg’s “GPO volume - Frankfurt exchange” function and found that the activity was very low. For example, for each of the three, five and ten year maturity GGBs there were less than EUR 0.5 million in published trades.

3.Broker quotes: In order to assess whether there was liquidity outside Greece for the Greek Government Bonds, we considered the bid/ask quotes by international brokers BGN, RBS Markets, Nomura, Caja Rural Castello and ING. The information provided by them indicated a very significant bid/ask spread even when compared to the spreads at 31 December 2010. Although we could not ascertain, by reading out to Bloomberg, if these quotes were executable or not, we believed that the wide bid/ask spread also supported our assessment that there was no liquidity in the markets for GGBs (ASC 820-10-35-51A (f)).  For example, for the three, five and ten year maturity bonds, these were the indicative bid/ask prices at 30 December 2011:

	31 December 2011
	Bid	Ask
GR0124021552 (3 years bond)
BGN	26.30	30.20
RBSM	15.75	38.25
Nomura	28.00	34.00
ING	27.42	29.42
GR0114023485 (5 years bond)
BGN	22.51	26.50
Caja Rural Castello	22.55	26.48
Nomura	21.00	27.00
ING	23.50	25.50
GR0124032666 (10 years bond)
BGN	19.00	25.00
RBSM	22.39	26.61
Nomura	19.00	25.00
ING	21.48	23.98
Reuters	22.26	28.21

29

As a reference point the bid/ask prices quoted by BGN in 31 December 2010 were:

	31 December 2010
ISIN/Maturity	Bid	Ask
GR0124021552 (3 years bond)	82.11	82.24
GR0114023485 (5 years bond)	75.46	75.58
GR0124032666 (10 years bond)	66.36	66.48

Note that the price for the GGBs implied by the discount rate of 12% that we used was 25.81 for all bonds as we priced our portfolio taking into account the haircut and terms of the PSI as well as the fact that all bonds would be replaced by the same new bonds. Therefore we believe the price we used was within the range of bid/ask quotes provided by these brokers.

4.In addition, we used the broker quotes above to support the conclusion that the market was inactive based on the guidance in ASC 820-10-35-51A (c) “price quotations vary substantially either over time or among market makers (for example, some brokered markets)”, (e) “there is a significant increase in implied liquidity risk premiums, yields, or performance indicators (…)”.

5.Furthermore, as the Greek government had been locked out of the international markets for bond issuances, we believe the indicator in ASC 820-10-35-51A(g) “there is a significant decline or absence of a market for new issuances (that is, a primary market) for the asset or liability or similar assets or liabilities” also supports our conclusion.

Therefore, we considered that the evidence we had that no active market existed based on the absence of trades in the HDAT and Frankfurt exchanges, and the large bid/ask spreads provided by major international brokers outweighed the information provided by the quotes available on Bloomberg.

On balancing all available evidence, we evaluated ASC 820-10-35-51B and concluded that the transactions or quoted prices (including the bid and ask prices) were not determinative of fair value and given the severity of the fact pattern (that is no trades in Greece and significantly high bid/ask spreads by international brokers), we determined that it was not necessary to perform any further analysis of the prices published by Bloomberg or the quoted bid/ask prices.

30

Based on the significance and relevance of the inexistence of actual trades in the Greek and Frankfurt bond markets, the high bid/ask spreads in both Greek and international markets and the deterioration of the Greek government economic conditions which effectively locked it out of international markets for new issuances, we believed that there has been a significant decrease in the volume and level of activity for the GGBs at December 31, 2011.

·Further, tell us how you determined it was appropriate to reclassify the GGBs from Level 2 to Level 3. Tell us how you considered the guidance in ASC 820-10-35-47 to 51.

We determined that it was appropriate to transfer the GGBs from Level 2 to Level 3 because the 12% discount rate used was determined using significant unobservable inputs as described in more detail in our response to the bullet point below.

·Tell us how you determined the discount rate of 12% was appropriate, and identify the guidance on which you relied.

As there has been a significant reduction in volume of transactions in GGBs, we relied on the guidance in paragraphs ASC 820-10-35-33(a) and the guidance beginning on ASC 820-10-55-4 which deal with present value techniques to estimate fair values. The paragraph states in part:

“… This Section neither prescribes the use of one specific present value technique nor limits the use of present value techniques to measure fair value to the techniques discussed herein. The present value technique used to measure fair value will depend on facts and circumstances specific to the asset or liability being measured (for example, whether comparable assets or liabilities can be observed in the market) and the availability of sufficient data.”

We believe our present value technique incorporates all requirements in ASC 820-10-55-5 as a willing market participant would. Specifically, we considered for our GGBs:

a.an estimate of future cash flows for the asset or liability being measured;

b.expectations about possible variations in the amount and/or timing of the cash flows representing the uncertainty inherent in the cash flows as a result of the PSI;

c.the time value of money, represented by the rate on risk-free monetary assets that have maturity dates or durations that coincide with the period covered by the cash flows (risk-free interest rate); and

d.the price for bearing the uncertainty inherent in the cash flows (risk premium)

In order to maximize the use of market-related inputs, we derived our discount rate from other bonds. More specifically, the discount rate of 12% was derived from the average yield of sovereign bonds with similar credit ratings to Greece.

31

The comparison was made against the yield of countries rated B- by at least two rating agencies. As of the date of the internal analysis, Fitch rated the Greek sovereigns at B-, while S&P rated it at CCC. The analysis was based on the fact that similarly-rated countries are expected to have similar fundamentals and overall economic conditions according to credit rating agencies’ methodology. (i.e. Republic of Belarus, Republic of Equador, Jamaica and Islamic Republic of Pakistan). The conclusion of the analysis was that cross-country comparables regarding maturity (20 years) and credit rating (B-: the lowest available) resulted to a yield within a range of 7.9%-13.1% and a mid-point estimate of 10.5%. Adjusting for Greece’s slightly lower rating (B- to CCC) the rate was estimated to 11%.

An additional risk premium related to Greece was applied, thus increasing the rate to 12%. On the basis of the significance of the unobservable inputs used in these methodologies, as well as the assessment of the discount rate, management determined that the valuation technique resulted in a Level 3 measurement.

Greek bonds not eligible for the PSI, page F-49

21.Your disclosure indicates that you have a recognized unrealized loss of Euro 27.4 million in other comprehensive income related to your GGBs maturing in May 2014. Your disclosure states that the fair value of these GGBs was determined by an independent appraiser as the bond is not quoted in an active market taking into account the credit risk of the Hellenic Republic. We do not believe the fair value determined for these bonds as of December 31, 2011 was reasonable because the valuation has not taken into consideration all available market information. Additionally, it does not appear that the bonds were appropriately valued on the date(s) of receipt. Specifically, we believe these bonds should have been recognized at fair value on the date(s) they were received in connection with Law 3273/2008 (Pillar 1), and it appears the value would be below the nominal value, and thus would also have implications for the value the preference shares were recorded at on the date of receipt. Please respond to the following:

·Tell us the dates you received these GGBs, and quantify the value you recorded for these GGBs on those dates.

We advise the Staff that the dates we received this GGB (the “May 2014 GGB”) and the value at which we recorded it, are as follows:

1st tranche, May 2009:

NBG received the first tranche of the May 2014 GGB on May 14, 2009 as a contribution in kind with respect to the share capital increase completed on the same date in the context of Pillar I of the Hellenic Republic Bank Support Plan. The nominal amount of this first tranche received by NBG in May 2009 was EUR 350 million and it was recorded

32

at its fair value, which, at that time, it was determined to be equal to the nominal value, that is the issue price was determined to be 100% (please refer to our response to the next bullet point for a discussion on how the fair value was determined).

2nd tranche, December 2011:

NBG received the second tranche of this bond on December 29, 2011 as a contribution in kind with respect to the share capital increase completed on the same date in the context of Pillar I of the Hellenic Republic Bank Support Plan. The nominal amount of this second tranche received by NBG on December 29, 2011 was EUR 1,084.7 million and it was recorded at its fair value, which, at that time, it was determined to be 92.19% of the nominal amount, that is EUR 1.0 billion (please refer to our response to the next bullet point for a discussion on how the fair value was determined).

·To the extent that these GGBs were recorded at their nominal value, instead of fair value, on the date received please tell us why. As part of your response, please tell us the fair value assumptions you used, or would have used, if fair value was determined on the date(s) received.

We advise the Staff that the May 2014 GGB was recorded at fair value based on ASC 820 on each of the dates received. The fair value was determined by the Bank as follows:

1st tranche, May 2009:

The May 2014 GGB is a 5 year floating rate bond with the coupon set at 6-month Euribor + 130 basis points (“bps”). The total nominal value of this bond issued in May 2009 was €3,690.3 million and was allocated to eight Greek banks that applied for support, based on the nominal amount of preference shares issued.

The terms of the May 2014 GGB were determined by the Hellenic Republic’s Public Debt Management Agency (the “PDMA”), by reference to a comparable GGB (the “reference bond”) that had similar characteristics to the May 2014 GGB, such that the issue price of the May 2014 GGB in May 2009 to be 100%. The table below presents the key terms and characteristics of the May 2014 GGB and the reference bond.

	May 2014 GGB	Reference GGB
Maturity:	May 21, 2014	May 20, 2014
Coupon:	6-month Euribor + 130 bps	4.5%, fixed
Total nominal amount:	EUR 3.7 billion	EUR 8.5 billion
Ratings	A1 by Moody’s, A- by S&P, A by Fitch and AA by R&I	A1 by Moody’s, A- by S&P and A by Fitch

33

Based on the market yield of the reference bond, which at that time was publicly traded in HDAT and the markets were active (see volume information in question 20 for the HDAT market), the PDMA determined that a market coupon rate (i.e. a rate that would imply an issue price of 100%) for the May 2014 GGB would be 6-month Euribor + 130 bps. Therefore, the contractual terms of the May 2014 GGBs were determined based on market prices and that is the reason we recorded it at par which was equal to its fair value. We advise the Staff that at the time of this issuance of the May 2014 GGBs, in May 2009, the Reference GGB was trading at 102.4% of its face value with an implied rate equal to 6-month Euribor + 130bps (6-month Euribor at the issuance was 1.465%).

In addition, a valuation committee (the “2009 Committee”), consisting of two qualified appraisers, was appointed in May 2009 by the Greek Ministry of Finance to value the May 2014 GGB. The 2009 Committee, as well as the Bank’s management, took into consideration the terms and characteristics of the reference bond, the volume of transactions for the reference bond, the fact that HDAT prices were used and the calculation performed by the PDMA, and concluded that the spread determined at 130 bps reflected the market conditions (at that time), hence the issue price of the May 2014 GGB at 100% of par value reflected the fair value of the bond.

2nd tranche, December 2011:

The total nominal value of the second tranche issued in December 2011 was €1,496.9 million and was used for the subscription of the additional preference shares of nominal value of €1,380.0 million, issued by NBG (EUR 1.0 billion) and another Greek bank (EUR 380.0 million).

The issue price of the 2nd tranche of the May 2014 GGB, 92.19% was determined by the PDMA, based on the terms of the May 2014 GGB (maturity: May 21, 2014, coupon rate: 6-month Euribor + 130 bps) and a discount rate of 6.5%.

As described in our response to question 20, there was not an active market for GGBs in the whole month of December 2011 in Greece or in international markets which we could use to derive the price of the May 2014 GGBs by comparing it to the quoted price of the reference GGB (there were nil transactions in HDAT for the reference GGB in December 2011) and the same valuation technique used for the other GGBs was applied by Management for the May 2014 GGBs with the differences described below.

In accordance with ASC 820-10-35-9 we evaluated which assumptions a market participant would take into account when valuing these bonds considering that the May 2014 GGBs were a private placement and therefore not traded and the market for GGBs in general was inactive during December 2011 (please refer to our response to question 20). We established that the key assumptions in determining this issue price:

(a)the contemplated (at that time) PSI would be successfully completed, and

(b)the May 2014 GGB would not be included in the contemplated PSI,

34

were assumptions which a market participant would have taken into account at that time because it was a common market expectation that the PSI would be implemented resulting in a significant discount on the face value of the PSI GGBs and that this particular bond was not included in the PSI as announced on February 24, 2012.

These assumptions are relevant because they differentiate the May 2014 GGBs from the other GGBs participating in the PSI which carried a known significant discount to par value as a result of the PSI agreed terms (on average approximately 64%, for the GGBs subject to Greek law held by the Group at December 31, 2011 and based on HDAT quotes).

Management then adopted the same methodology to determine the fair value of the May 2014 GGBs as described in question 20 with the exception that instead of referring to countries rated B- by at least two rating agencies (Republic of Belarus, Republic of Ecuador, Jamaica and Islamic Republic of Pakistan), management referred to European Union countries which were also under distress but that have not implemented a PSI plan in order to appropriately incorporate in the valuation the fact the May 2014 GGBs would not be in the PSI and therefore would be entitled to its full contractual cash flows at maturity date in May 2014 (we believe this is consistent with the requirements in ASC 820-10-55-10).

The price was based on an analysis of the average yields during December 2011 of Italian sovereign bonds with similar maturity, Italy being the only Eurozone country with public debt greater than 100% similarly to Greece that still has access to the bond market and had not participated in any debt reduction plans with its bondholders. In addition we also considered that Italy had a similar debt/GDP ratio as Greece (after considering the benefit of the PSI). Therefore, we considered that the rates implied in Italian bonds were the best proxy for a discount rate a market participant would apply to the May 2014 Bonds in accordance with ASC 820-10-55-10.

This approach and the resulting price was also consistent with the approach taken by a Committee appointed by the Hellenic Republic to determine the price range for the May 2014 GGB.

Lastly, management acknowledged that the economic situation in Greece was different than in Italy and therefore a higher discount rate might have been appropriate. Management decided not to use a higher discount rate for the following reasons:

(a)all GGBs that were either not included in the PSI (held by ECB and other central banks) or included in the PSI but not tendered (hold-out bonds) with total nominal value EUR 5.1 billion and matured between March 12, 2012 (the date of the first PSI exchange) and May 15, 2012 (the date of filing of our Form 20-F) were repaid in full and on time.

(b)after the completion of the PSI, but before the US GAAP financial statements were issued, we estimated May 2014 GGB’s price by reference to:

–the market yields of the new GGBs issued in the context of the PSI, in order to take into consideration market observable inputs that incorporate the impact of the PSI, and

35

–the market yields of Greek treasury bills, which have been issued with certain frequency by the Greek government since the PSI was implemented. We used these treasury bills in order to construct the short-end of the zero coupon curve used to discount the cash flows of the May 2014 GGB, since the shortest maturity of the new GGBs is February 2023.

This exercise resulted in an estimated price of 89.2% at March 30, 2012. Management believed that based on the methodology used which determined the 92.19% and the immaterial difference between the 89.2% and the 92.19%, the assumptions used at December 31, 2011 were appropriate.

We also considered whether our price was within the acceptable price ranges identified by two separate valuation committees appointed by the Hellenic Republic and the Bank, respectively. The committee appointed by the Bank was supported by PwC whose report identified a range of acceptable values between 79.2 and 94.3 and we concluded that the price estimated as described above was within the range of all valuation reports.

Based on all the above evidence, the Bank concluded that the issue price of 92.19% reflected the fair value of the May 2014 GGB as of December 29, 2011 (the date of receipt) and as of December 31, 2011 (the date of the financial statements), in accordance with ASC 820. Furthermore, the fair value of both the 1st and the 2nd tranches of the May 2014 GGB, at December 31, 2011 was calculated using the price of 92.19%.

We also advise the Staff that performing the valuation exercise described in (b) above using September 25, 2012 market yields for the new GGBs and treasury bills, results in a price of 92.13%, that is very close to the price used at December 29, 2011.

·Please revise your fair value determinations as of December 31, 2011 and as of the date of receipt of the bonds to take into consideration all available market information existing as of the date(s) of valuation.

We respectfully advise the Staff that we continue to believe that our fair value determinations as of December 31, 2011 and as of the date of receipt of the May 2014 GGB, as described in our response to the bullet point above, did take into consideration all available market information existing as of the date of valuation. Therefore, we believe it would not be appropriate to revise our fair value determinations as of December 31, 2011 with respect to the May 2014 GGB.

36

22.We are also unable to agree with your conclusion that these May 2014 GGBs are not other-than-temporarily impaired as of December 31, 2011. Specifically, your analysis regarding the lack of other-than-temporary impairment is based on the presumption that your Greek banking regulator will allow you to reduce your capital by exchanging the preference shares or that the Greek government will have sufficient funds to repay the bonds in May 2014. Given the uncertainties existing in each of the scenarios, it is still unclear to us how you concluded that these GGBs should not be other-than-temporarily-impaired.

We advise the Staff that our OTTI analysis with respect to the May 2014 GGB took into consideration all the evidence presented in our response to comment 21 above, as well as:

(i)The fact the Bank did not intend to sell the securities and that it was not more likely than not that the Bank would be required to sell the securities.

(ii)The insignificant unrealized losses of EUR 27.4 million, representing 2% of the amortized cost basis (refer to our response to your comment 21 above for a discussion on how the fair value for this particular bond was determined at December 31, 2011).

(iii)The reduced risk of non-payment compared to the other GGBs that were included in the PSI for the reasons described on page F-50 of our Form 20-F and in our response to your comment 21 above. We do not consider that there are such uncertainties regarding the ability of the Hellenic Republic to repay this bond that, when considered together with the insignificant unrealized losses, would lead to the conclusion that this bond is other-than-temporarily-impaired.

(iv)The fact that all GGBs that were not tendered in the PSI (held by ECB and other central banks or hold-out bonds) with total nominal value EUR 5.1 billion and matured between March 12, 2012 (the date of the first PSI exchange) and May 15, 2012 (the date of filing of our Form 20-F) were repaid in full and in time.

We also advise the Staff that all GGBs that were not tendered in the PSI (held by ECB and other central banks or hold-out bonds) with total nominal value EUR 11.8 billion and matured between March 12, 2012 (the date of the first PSI exchange) and September 28, 2012 (the date of this letter) were repaid in full and in time.

23.We note your disclosure on pages F-50 and F-51 related to your impairment analysis on the bond issued by the New Economy Development Fund (TANEO), a company controlled by the Hellenic Republic, which has a nominal value of Euro 57.7 million and a carrying value of Euro 56.0 million at December 31, 2011. You state that, because TANEO holds cash and cash equivalents in an amount that exceeds the total bond issue, you believe the probability of TANEO defaulting on the bonds is very low. Please respond to the following:

·Tell us the business purpose of the TANEO entity and clarify how the Hellenic Republic controls TANEO. For example, does the Hellenic Republic own all of the equity of the entity or are there other parties involved in the ownership structure?

37

We advise the Staff that, based on publicly available information, the business purpose of TANEO is the minority participation in closed-end venture capital funds (AKES), venture capital companies (EKES) and similar venture capital organizations (hereafter referred to as “investment organizations”), which will be established specifically for this purpose and will operate in accordance with the laws of any Member State of the European Union. These investment organizations must be managed by private entities in conformity with private sector financial criteria and must invest exclusively in innovative small and medium-size enterprises in Greece.

The share capital of TANEO as at 31.12.11 amounts to EUR 46.0 million, which comprises of:

·EUR 1.0 million ordinary share capital represented by 20,000 common shares with voting rights of par value EUR 50 each, and

·EUR 45.0 million preferred stock, represented by 900,000 preferred shares of par value EUR 50 each.

The Hellenic Republic owns all the common and preferred shares issued by TANEO.

·Tell us whether TANEO has any other issued bonds, and, if so, how this is factored into your other-than-temporary analysis on this bond. To the extent that the entity has issued other bonds, please tell us the maturity dates of the other issuances and whether any are senior to the bonds held by you.

TANEO only has this bond issue with a nominal value of EUR 105.0 million as of December 31, 2011 and a maturity of June 3, 2013, in which the Bank and other private investors have invested. TANEO has no other issued bonds.

·Tell us whether TANEO is contractually restricted from using the cash and cash equivalents held by it for other purposes besides repaying the outstanding bonds you hold. For example, is the cash held in a trust?

According to TANEO’s IFRS 31.12.11 financials the company has cash & cash equivalents of EUR 1.4 million and Level 1 assets valued at EUR 100.5 million of which EUR 97 million are money market funds and EUR 3.5 million are capital funds. The Level 3 assets are valued at EUR 27.6 million. TANEO’s liabilities comprise of this bond whose total amount in issue is EUR 105.0 million and the preferred stock of EUR 45.0 million, which in the IFRS financial statements is presented as a liability and ranks subordinate to the bond. Considering that the bond has 18 months to maturity and there is 101.9 million in Level 1 assets and cash & cash equivalents we concluded that as of December 31, 2011 TANEO will be able to repay its bondholders EUR 105 million on the maturity date June 3, 2013. The remaining liabilities of TANEO as of December 31, 2011 amounted to EUR 1.5 million and thus are immaterial.

38

We would like to inform the staff that there is a cash management account as described in the audited financial statements of TANEO for the year ended 31.12.11.

“In line with the terms of the information memorandum and the documentation signed as part of the bond issue of €105m, any cash in the accounts maintained by the company with Deutsche Bank AG London may be invested in Eligible Obligations. Eligible Obligations means money market funds and similar investment schemes expressed in euro with a credit rating equal to AAA. Consequently, the securities held by the company on 31/12/2011 amounted to €97m and related to cash invested in the Deutsche Global Liquidity Series — Deutsche Euro Managed Fund”.

This cash management account is for the benefit of the bondholders as stated in the prospectus:

“The Hellenic Republic will agree in the Cash Management Agreement not to receive any payments in respect of the Dividends or the redemption of the Preferred Shares Outstanding Balance before the date on which the Noteholders have been paid in full in respect of the Principal Balance and all Guaranteed Interest accrued to that date in respect of the Notes.  Following payment in full of the Principal Balance and the accrued Guaranteed Interest in respect of the Notes, payment of any Dividends or the redemption of the Preferred Shares Outstanding Balance may be made subject to sufficient funds being available in accordance with Condition 3 (Priority of Payments and Available Funds) of the Notes”.

·Given that TANEO is controlled by the Hellenic Republic, tell us how you concluded that they are unable to use the cash held by this entity for other purposes.

The Hellenic Republic is a shareholder in TANEO and has the rights and obligations of any shareholder under Greek Law and thus would be unable under the current law to use TANEO’s money market investments for other purposes (refer also to our response in the previous bullet).

Taking all that into account, the Company concluded that this investment is not other-than-temporarily impaired.

Note 13: Loans and Allowance for Loan Losses, page F-53
Allowance for loan losses, page F-59

24.Your table on page F-60 quantifies your various loans to the Greek state and related exposures. While the allowance for your loans eligible to PSI is approximately 74% of the loan balance, the allowance for loans that were not eligible for the PSI exchange is only approximately 7% of the loan and related exposure balance. Please more clearly disclose how you considered the impact of the credit losses of the PSI exchange in determining the allowance for these other loans and exposures.

39

We advise the Staff that we will add the following disclosure in our next Annual Report on Form 20-F:

“The allowance for loan losses on PSI eligible loans was estimated at approximately 74% of the carrying amount of the said loans based on the terms and conditions of the Offer as described in Note 11 as it represented incurred known losses. The 74% loss rate also formed the initial basis for our allowance for loan loss calculation for the other exposures to the Hellenic Republic presented in the above table. Our calculation also took into consideration (a) the specific characteristics of each type of exposure and (b) the fact that there was no specific loss event in relation to these exposures as was the PSI for the PSI eligible loans, (c) the fact that all these exposures with the Hellenic Republic were being serviced, and (d) the relief promoted in the other obligations of the Hellenic Republic by the PSI and the financial support provided by the Troika.

More specifically, for direct exposures to the Hellenic Republic, the loss rate was adjusted downwards by applying a probability that the Hellenic Republic will default on the particular exposure after considering all the factors described above.

For the exposures to other entities, guaranteed by the Hellenic Republic, the loss assessment first considered the ability of the borrower to repay the loan, using our normal methodology as described in Note 3. For the cases that we determined that repayment of the loan depends in whole or in part to the guarantee provided by the Hellenic Republic, the 74% PSI loss rate was applied, adjusted downwards by an estimate of the probability that the Hellenic Republic will default after considering all the factors described above.

For the mortgage loans guaranteed by the Hellenic Republic presented in the table above, we use our normal methodology as described in Note 3.”

Economic conditions and trends, page F-62

25.You state that the deterioration in economic conditions was reflected in your allowance for loan losses in the three ways identified here. The first and third ways that you identified here for considering these adverse economic conditions seem to be simply the exercise of your normal allowance methodology in dealing with increasing past due loans and declines in real estate collateral values, whereas the second way your methodology considered these adverse economic conditions was to increase the probability of default for computing your incurred but not identified losses in your homogeneous portfolios. Please identify the period or periods in which you increased the probability of default for computing your incurred but not identified losses, and quantify the impact on your allowance and provision for loan losses.

40

We advise the Staff that the probability of default used in the calculation of the allowance for loan loss increased both in 2011 and 2010 due to the adverse economic conditions and the deterioration of the quality of our loan portfolio evidenced, in part, by the higher frequency of default observed. Had we used the probability of default used in 2010 to compute domestic allowance for loan losses in 2011, the allowance for loan losses as well as the provision for loans losses would have been lower by EUR 39.6 million.

26.Your disclosure further notes your conclusion that no further adjustment to your historical loss rates was necessary because the loss rates applied are long term averages based on long recovery periods. Please explain what you mean here, and discuss the extent to which the historical periods used to derive your loss rates are and are not reflective of the current macroeconomic environment. Identify the limitations of using historical loss rates to estimate incurred but not yet identified losses in periods which are markedly dissimilar to current and future economic environments. To the extent that you are using recovery rates from historic periods, discuss how your model considers the fact that the current macroeconomic environment can be expected to produce significantly different recoveries that those of the last six years. Discuss how you addressed these limitations in your allowance methodology, beyond your increases to the estimated probability of default as noted above.

We advise the Staff that “long recovery periods” in the loss rates calculation refers to the assumed recovery periods of six years for consumer loans and credit cards and twelve years for mortgages for the purposes of calculating our homogeneous loan loss allowances. As disclosed in Note 3 on page F-19 of our Form 20-F for 2011, recovery periods in Greece can be very long due to existing legislation, the judicial system and business practices.

Furthermore, “long term averages” refer to method of estimating the average loss rates used in our homogeneous loan loss allowance. As described in our response to your comment 11, in our letter dated December 9, 2011, we try to incorporate in our analysis as many 6 year recovery periods as possible. In particular, for the December 31, 2011 calculation we included recoveries data starting with loans that defaulted in 1999, based on which we calculated weighted average loss rates (based on the number of individual accounts that default each year).

The recent historical periods used, that is recoveries since 2008 (the first year of recession in Greece), and in particular the recoveries data since 2010 when the Greek economic crisis exacerbated, are reflective of the current macroeconomic environment. The recoveries data from periods prior to 2008, that is, periods with economic growth and better loan portfolio quality compared to the post 2008 period, is not reflective of current macroeconomic environment.

Using historical loss rates to estimate incurred losses in the existing loan portfolios has the inherent limitation that the past data may not be reflective of current conditions. This limitation becomes even more prominent when current economic conditions are significantly different than the conditions existing during the periods used to derive the loss rates.

41

As described in our response to your comment 11, in our letter dated December 9, 2011, our model considers the above limitation by assigning more weight to the recoveries data of the more recent years ( 2008-2011), being the years with the higher number of defaults.

Therefore, we concluded that current economic conditions are fairly considered and as at December 31, 2011 we did not adjust our loan loss allowance methodology. However, we did update the estimates for the parameters used in our methodology by increasing the estimated probability of default and the estimated loss rates, as described above.

Note 29: Other Non-Interest Expense, page F-92

27.Please revise your footnote or MD&A as appropriate to more clearly discuss the amounts reflected in the line item titled “Commissions on issues of Hellenic Republic Bank Support Plan — Pillar II.” Clearly identify how such amounts are determined.

We advise the Staff that these commissions refer to guarantee fees paid by the Bank and in our next Annual Report on Form 20-F, we will amend our disclosure on pages F-73 and F-92, as follows:

Page F-73, below the table:

“The notes described above have been issued by the Bank, are guaranteed by the Hellenic Republic and are held by the Bank and therefore, are not presented as liabilities on the Consolidated Balance sheet since the notes are not outstanding. Additionally, the notes are currently rated · by Moody’s and · by Fitch and can be used by the Bank in repo transactions with the ECB. For the guarantees provided by the Hellenic Republic, an annual fee is payable by the Bank to the Hellenic Republic. This fee ranges from 95 to 135 basis points (depending on the notes) and is calculated on the nominal amount of the notes issued, pro rated for the period the notes are outstanding.”

Page F-92, add the following disclosure below the first table:

“Commissions on issues of Hellenic Republic Bank Support Plan — Pillar II relate to the fees payable to the Hellenic Republic with respect to the notes issued by the Bank and guaranteed by the Hellenic Republic, in the context of Pillar II of the Hellenic Republic Bank Support Plan, as described in Note 17.”

42

Should you have any questions or require any additional information, please contact Mr. Panagiotis Chatziantoniou, Head of Finance Division, on + 30 210 3695708.

For the National Bank of Greece S.A.
Yours sincerely,

/s/ Petros Christodoulou

Petros Christodoulou

Deputy Chief Executive Officer

Copy:	Manos Pelidis, Deloitte Hadjipavlou Sofianos & Cambanis S.A.
	Diana Billik, Allen & Overy LLP
	Alan L. Beller, Cleary Gottlieb Steen & Hamilton LLP

43