Banco Latinoamericano de Comercio Exterior, S. A. (BLX) CORRESP Correspondence with SEC

November 22, 2006

Mr. Kevin W. Vaughn
Accounting Branch Chief
Division of Corporation Finance
United States Securities and Exchange Commission
Washington, D.C. 20549

Re:	Latin American Export Bank
	Form 20-F for the Fiscal Year Ended December 31, 2005
	File No. 001-11414

Dear Mr. Vaughn,

          As indicated in our letter dated November 7, 2006, please find our responses to the comments raised in the Staff’s comment letter dated September 19, 2006 regarding the Bank’s Form 20-F for the fiscal year ended December 31, 2005.  The Staff’s comments are repeated in bold and italics with the Bank’s responses noted below each comment.

Background

We thought it would be helpful to provide some background concerning the factors that have affected the Bank’s allowance for credit losses during recent years.  We have discussed these factors in detail in the Bank’s Forms 20-F for the fiscal years ended December 31, 2001 through 2005, as well as in the Bank’s quarterly reports filed under the cover of Forms 6-K during the period from 2001 to 2005. 

During the period from 2003 to 2005, the two principal factors affecting the Bank’s allowance for credit losses were:

1)	The collection of the impaired Argentine portfolio, which led to US$227 million in reversals of specific allowance for credit losses assigned to this Argentine portfolio (please see Annex 1); and
2)	The implementation of a new methodology for assessing the Bank’s generic allowance for credit losses during the third quarter of 2005, which, although important, had a much smaller impact than the latter (a net effect of US$7 million provision of allowance for credit losses). Please see Annex 1 and Annex 3.

Comment responses

Investment Securities, page F-14

1.

We note that a material portion of your investment portfolio is comprised of various government securities.  In order to provide the user of the financial statements with a better understanding of the relative risks associated with these investments, please revise your future filings to provide a tabular breakdown that identifies the government issuers of whom you hold significant balances in your portfolio.

We have provided in Annex 2 a more detailed presentation of the Bank’s investment portfolio.  We confirm to the Commission that we would provide this information in future filings.

Allowance for Credit Losses, page F-16

2.	Please address the following regarding your disclosure that during the third quarter of 2005, you implemented a new methodology for estimating generic allowance for credit losses which incorporates the Bank’s historical loss performance in the calculation of expected losses rather than using default data from rating agencies.
	•	Please tell us in detail how you determined that your use of an expected loss model is in accordance with U.S. GAAP, including the specific authoritative literature relied upon in making your determination.

1

	Bladex developed an expected loss model for the generic allowance for credit losses based on the guidelines provided by FAS 5, Accounting for Contingencies.  FAS 5 establishes standards of financial accounting and reporting for loss contingencies by:
		1.	Accruing an estimated loss from a loss contingency by a charge to income, if both of these conditions are met: (i) information available prior to issuance of the financial statements indicates that it is probable that an asset had been impaired or a liability had been incurred at the date of the financial statements; and (ii) the amount of loss can be reasonably estimated. (FAS 5, par. 8);
		2.	Whether the amount of loss can be reasonably estimated will normally depend on, among other things: (i) the experience of the enterprise, information about the ability of individual debtors to pay, and (ii) the appraisal of the receivables in light of the current economic environment. (FAS 5, par. 23); and
		3.	In the case of an enterprise that has no experience of its own, reference to the experiences of other enterprises in the same business may be appropriate (FAS 5, par. 23).
	We believe that the expected loss model used by Bladex to assess its generic allowance for credit losses is in line with FAS 5 because it provides a reasonable estimate of credit losses by calculating Probability of Default and Loss Given Default factors (FAS 5, par 8), supported by our own historical experience (FAS 5, par. 23).
•	For both the allowance for loan losses and the reserve for losses on off-balance sheet credit risk, clearly describe to us and in your future filings how you used an expected loss model in arriving at your allowances balances.
	The allowance/reserve balances, for both on and off-balance sheet exposures are calculated applying the following formula:
		Allowance/Reserve = Exposure (a) x Probability of Default (b) x Loss Given Default (c), where:
			a)	Exposure = the total accounting balance (on and off-balance sheet) at the end of the period under review, segregated by country.
			b)	Probability of Default (PD) = Probability of Default of the corresponding country classification, supported on Bladex’s historical portfolio performance for an eight-year period, and complemented by S&P probabilities of default for higher risk cases, in view of the greater robustness of S&P data for such cases. These probabilities of default factors are applied to each country. Management judgment applies in determining each country risk classification.
			c)	Loss Given Default (Expected Losses) = a factor of 45% is utilized, based on best practices in the banking industry. This factor applies to all countries, except when a country is classified as Higher Risk, in which case management judgment is applied in estimating the Loss Given Default on a case-by-case basis.
•	For both the allowance for loan losses and the reserve for losses on off-balance sheet credit risk, clearly identify any adjustments to the expected loss model outputs that you made as part of your use of this data.
	No adjustments to the model, as described in the previous response, are involved.
•	Explain in detail, why the change in methodology resulted in a positive impact on 2005 net income as it relates to the allowance for loan losses and a negative impact as it relates to the reserve for losses on off-balance sheet credit risk.
	•	The positive impact on 2005 net income as it relates to the allowance for loan losses results from the use by the Bank of its own historical portfolio performance to determine the Probabilities of Default, whereas the previous methodology utilized only S&P Probabilities of Default, which are more severe than the ones applicable to the more selective and specialized nature of the Bank’s portfolio (short-term trade finance).  To a lesser extent, the result was also affected by the fact that the new methodology utilizes one-year probabilities of default, given the short-term nature of the Bank’s portfolio (average maturity of 264 days), whereas the previous methodology utilized the probabilities of default over the remaining tenor of each loan, which resulted in more severe factors when exposures were longer term.

2

		•	With regard to the reserve for losses on off-balance sheet credit risk, the negative impact in 2005 resulted from the fact that the previous methodology considered a lower probability of default for off-balance sheet items, whereas the new methodology applies the same factor to both on and off-balance sheet items. This is because the Bank’s data at this date is not sufficient to allow for segregated probabilities of default on a robust basis.  (In this regard, we note that the Bank was forced in the aftermath of the Argentine crisis to reserve for both on and off-balance sheet items on an equal basis).
	•	Tell us why you believe this qualifies as a change in accounting principle from one acceptable method to another under paragraph 7-8 of APB 20.
		As indicated in previous responses, the new methodology is now driven mainly by Bladex’s own historical probability of default and loss experience, complemented by external data, rather than relying exclusively on third party average market data, as was formerly the case.  In addition, the new methodology incorporates management’s judgment included in the Probability of Default and the Loss Given Default factors through the country risk classifications.  These factors constitute a substantial change in methodology with respect to the previous methodology, and as a result Bladex believes that it qualifies as a change in accounting principle from one acceptable method to another under paragraph 7-8 of APB 20.
3.	As a related matter, please revise to disclose:
	•	More clearly discuss the reasons for the change in methodology including the underlying reasons for changing in the current year as opposed to changing in a prior period;
		The change in methodology was the result of the Bank’s intention to follow best practices in the banking industry and is in line with FAS 5, which calls for internal historical performance data in the estimation of credit loss reserves.  The effort included consultations with some world-class financial institutions regarding their practices.  Please see Annex 3 for further information.
		Statistical data was not available until 2005. The Bank began compiling its historical data in 2004, completing the effort during 2005.
	•	Discuss whether you tracked your historical loss performance in prior years, and the extent to which you used that data in your allowance methodology in the past;
		As indicated in the previous response, no statistical compilation of data existed prior to the year 2005.
	•	If you tracked your historical data in prior years, discuss how you reconciled the default information from rating agencies with the historical losses experienced in your portfolio; and
		Not applicable; please see the responses above.
	•	More clearly disclose the events that occurred during each of the last several reporting periods that caused you to determine that a significant reversal was appropriate for those respective periods. Clearly identify the triggering events that occurred during each period that led to the reversal.
		The reversals of allowance for loan losses made during the last 3 years were primarily related to the collection of the impaired Argentine portfolio, which led to significant reversals of the specific allowance for loan losses that were assigned to each asset in this portfolio.  Please see Annex 1.
		During 2001 and 2002, the Bank increased the allowance for credit losses by US$77 million and US$279 million, respectively, bringing the total allowance for credit losses assigned to its Argentine portfolio to US$380 million at December 31, 2002.  As a result of payments and prepayments received, as well as write-offs, this allowance for credit losses had been reduced to US$21 million at the end of 2005.

3

No changes have been made in the methodology used to establish specific allowances/reserves, as the changes made in the allowance/reserve methodology apply only to the Bank’s generic allowance for credit losses.  The reversals of Argentine specific allowance for credit losses were thus not related or attributable to the changes in the reserve methodology.

The following table sets forth information regarding the Bank’s reversals (provisions) of allowance for loan losses during the years indicated:

Reversals (Provisions) of Allowance for Loan Losses

($ millions)	2003			2004			2005
Reversals of allowance for loan losses related to the sale of Argentine loans		62.8			6.3			2.9
Reversals of allowance for loan losses related to Argentine loan restructurings and collections		5.0			92.5			45.1
Total reversals of Argentine specific allowance for loan losses		67.8			98.8			47.9
Brazil reversals (provisions) of specific allowance for loan losses		2.3			(2.2	)		13.2
Total Reversals of Specific Allowance for Loan Losses		70.2			96.6			61.1
Reversals (provisions) of generic allowance for loan losses due to changes in loan portfolio composition and risk levels		(2.6	)		8.4			(15.5	)
Reversals of generic allowance for loan losses - due to change in loan loss reserve methodology		—			—			6.0
Total Reversals (Provisions) of Generic Allowance for Loan Losses		(2.6	)		8.4			(9.6	)
Recoveries - Argentine loans		1.1			6.4			0.3
Recoveries - Other		0.9			—			2.3
Total Recoveries		2.0			6.4			2.6
Total Reversals of Allowance for Loan Losses		69.5			111.4			54.2

1.	2005
	The Bank’s US$54 million reversal of provision for loan losses during 2005 was mainly due to the net effect of:
	•	a US$61 million reversal related to the decrease in restructured loans, reflecting payments and prepayments received during the year;
	•	a US$3 million reversal due to recoveries from loans charged-off in previous years;
	•	US$16 million in generic provision charges, due to increased loan exposure; and
	•	a US$6 million reversal due to the change in the credit loss reserve methodology during 2005.
2.	2004
	The US$111 million provision reversal during 2004 was mainly due to reversals of provisions related to Argentina, including:
	•	US$92 million in payments and prepayments received on restructured loans,
	•	US$6 million in loan recoveries, and
	•	US$6 million in sales of loans.
	In addition, during 2004, the Bank decreased US$8 million in generic reserves assigned to certain countries mainly due to improved risk levels, and reversed US$2 million in specific reserves related to a non-accruing loan in Brazil due to payments received.  These reversals were partially offset by a US$4 million increase in specific reserves assigned to a loan in Brazil placed on non-accrual status during the fourth quarter of 2004.

4

3.	2003
	The Bank’s US$70 million reversal of provision for loan losses during 2003 was mainly due to the net effect of:
	•	US$63 million reversal in specific reserves related to sales of Argentine loans,
	•	US$5 million reversal in specific reserves related to payments and prepayments received on restructured Argentine loans,
	•	US$2 million reversal in specific reserves related to a non-accruing loan in Brazil due to payments received,
	•	US$3 million in generic provision charges, due to changes in loan portfolio and risk levels,
	•	US$1 million reversal due to recoveries from Argentine loans charged-off in previous years; and
	•	US$1 million reversal due to recoveries from other loans charged-off in previous years.

5

          The Bank acknowledges that (i) the Bank is responsible for the adequacy and accuracy of the disclosure in the filing; (ii) staff comments or changes to disclosure in response to staff comments do not foreclose the Commission from taking any action with respect to the filing; and (iii) the Bank may not assert staff comments as a defense in any proceeding initiated by the Commission or any person under the federal securities laws of the United States.

          We look forward to discussing our responses and analysis with you once you have had an opportunity to review these materials.  In connection with all future filings, the Bank will provide the tabular information requested by the Staff’s comment noted above.  If you have any questions, please contact the undersigned at the following telephone number: Country Code: (507) phone number: 210-8561, e-mail: jarivera@blx.com.

Best regards,

/s/ Jaime Rivera

Jaime Rivera

Chief Executive Officer

6

Annex 1

Reversals of Argentine Specific Provision for Credit Losses 

The economic crisis in Argentina that began in December 2001 resulted in the imposition by the Argentine government of a number of measures, including a freeze on bank deposits, forced conversion of dollar-denominated bank deposits and the imposition of exchange controls.  As a result of the Argentine crisis, the Bank’s Argentine obligors faced repayment difficulties.  At December 31, 2001, the Bank’s Argentine credit portfolio totaled US $1.1 billion.  The Bank classified as impaired nearly its entire Argentine exposure due to these collectibility concerns and increased its allowance for credit losses during 2001 and 2002 by US $77 million and US $279 million, respectively, bringing the total credit reserves assigned to its Argentine portfolio to US$380 million at December 31, 2002.  Beginning in 2002, the Bank negotiated the restructuring of its Argentine portfolio and sold at a discount those positions that the Bank estimated had the lowest probability of collection.  As a result of the terms of its restructuring and within an improving scenario in Argentina resulting from higher commodity prices and the restructuring of its sovereign debt, the Bank in 2003, 2004 and 2005 reported loan loss provision reversals of US$70 million, US$111 million and US$54 million, respectively, of which US$69 million, US$105 million and US$48 million, respectively, were related to the reduction of the Argentine impaired loan portfolio, which decreased by US$294 million, US$191 million, and US$184 million for the years 2003, 2004 and 2005, respectively.  These reversals resulted from loan collections that exceeded their respective net book values.

Implementation of the new methodology for assessing the Generic Allowance for Credit Losses

In order to better align itself with best market practice for estimating generic reserves, the Bank reviewed its methodology for assessing the generic allowance for credit losses and decided to prepare itself for the implementation of an internal portfolio performance based approach.  During the third quarter of 2005, the Bank implemented a new methodology for assessing the generic allowance for credit losses.  The effect of the change in methodology during 2005 resulted in credit provision charges of US$10 million, represented by a US$6 million reversal of provision for loan losses and a US$16 million provision for losses on off-balance sheet credit risk.  The adoption of the methodology had an effect on years prior to December 31, 2004, resulting in a reversal of provisions for credit losses of US$3 million (consisting of a reversal of provision for loan losses of US$6 million and a provision for losses on off-balance sheet credit risk of US$3 million).

7

Annex 2

Banco Latinoamericano de Exportaciones, S. A.Investment Securities
	At December 31,
			2004				2005
(In thousands of US$)			Amortized Cost		Fair Value		Amortized Cost		Fair Value
Securities held to maturity
Government
Mexico		US$		27,984		28,117		26,520		26,325
Securities available for sale
Corporate
Argentina				665		932		—		—
Brasil		.		—		—		16,967		16,764
Chile				—		—		16,575		16,384
Panama				—		—		5,014		5,085
Sub-total				665		932		38,556		38,233
Government
Argentina				—		—		4,336		4,194
Brasil				11,036		11,387		80,285		80,808
Colombia				20,138		23,556		11,238		11,301
Chile				40,988		40,528		16,566		16,043
El Salvador				15,000		15,000		20,000		20,000
Mexico				73,438		73,467		11,711		11,470
Sub-total				160,599		163,939		144,136		143,817
Total securities available for sale				161,264		164,871		182,691		182,050
Total investment securities		US$		189,248		192,988		209,211		208,375

8

Annex 3

CONTEXT FOR DEVELOPMENT OF THE NEW METHODOLOGY

During its 27 years of operations, Bladex had never had to face a severe crisis until the 2001-2002 Argentine crisis and its aftermath. That crisis led the Bank to reserve nearly 50% of its exposure in the country, substantially impacting its net worth and leading the Bank to seek at least US$100 million in additional capital.  This led the Bank to consider the use of an improved methodology, which might be able to more closely reflect developments and risk in the Region. This is considered especially important in light of the substantial volatility, which affects the Region in which the Bank operates, both in terms of reacting on a more timely basis to negative developments, as well as in seeking additional business opportunities when and wherever conditions are considered to be favorable.

The use of internal probabilities of default to the extent possible, and the possibility of using the Bank’s extensive experience and knowledge of the Region were considered among the most important improvements to be made. The latter especially in view of the generally lagging nature of external, objective sources of information previously used to assign country risk ratings. It is noted that sudden and strong changes in country risk classifications are much more relevant in the Region than in more developed areas such as Europe or the United States, given the volatility, susceptibility to economic, political and social crisis, and the markedly different situations and vulnerabilities in countries which might otherwise appear to be similar.

As such, the move was made away from the previous methodology, which assigned reserves with various adjusting factors, to a simpler and more straightforward methodology based on probabilities of default and loss given default factors, incorporating management’s judgment with regard to country risk classifications. 

9