Arrow Financial - 10-Q quarterly report FY

1

ARROW FINANCIAL CORPORATION

FORM 10-Q

September 30, 2005

INDEX

2

ARROW FINANCIAL CORPORATION AND SUBSIDIARIES

CONSOLIDATED BALANCE SHEETS

(Dollars in Thousands) (Unaudited)

See Notes to Unaudited Consolidated Interim Financial Statements.

3

ARROW FINANCIAL CORPORATION AND SUBSIDIARIES

CONSOLIDATED STATEMENTS OF INCOME

(In Thousands, Except Per Share Amounts)(Unaudited)

Share and Per Share amounts have been restated for the September 2005 3% stock dividend.

See Notes to Unaudited Consolidated Interim Financial Statements.

4

ARROW FINANCIAL CORPORATION AND SUBSIDIARIES

CONSOLIDATED STATEMENTS OF CHANGES IN SHAREHOLDERS’ EQUITY

(In Thousands, Except Share and Per Share Amounts) (Unaudited)

Cash dividends declared have been adjusted for the September 2005 3% stock dividend.

Included in the shares issued for the 3% stock dividend in 2005 were treasury shares of 100,545 and unallocated ESOP shares of 2,436.

See Notes to Unaudited Consolidated Interim Financial Statements.

5

ARROW FINANCIAL CORPORATION AND SUBSIDIARIES

CONSOLIDATED STATEMENTS OF CHANGES IN SHAREHOLDERS’ EQUITY

(In Thousands, Except Share and Per Share Amounts) (Unaudited)

Share and Per Share amounts have been restated for the September 2005 3% stock dividend.

See Notes to Unaudited Consolidated Interim Financial Statements.

6

ARROW FINANCIAL CORPORATION AND SUBSIDIARIES

CONSOLIDATED STATEMENTS OF CASH FLOWS

(Dollars in Thousands)(Unaudited)

See Notes to Unaudited Consolidated Interim Financial Statements.

7

ARROW FINANCIAL CORPORATION AND SUBSIDIARIES

NOTES TO UNAUDITED CONSOLIDATED INTERIM FINANCIAL STATEMENTS

September 30, 2005

1.   Financial Statement Presentation   

In the opinion of the management of Arrow Financial Corporation (Arrow), the accompanying unaudited consolidated interim financial statements contain all of the adjustments necessary to present fairly the financial position as of September 30, 2005 and December 31, 2004; the results of operations for the three-month and nine-month periods ended September 30, 2005 and 2004; the changes in shareholders’ equity for the nine-month periods ended September 30, 2005 and 2004; and the cash flows for the nine-month periods ended September 30, 2005 and 2004.  All such adjustments are of a normal recurring nature.  The unaudited consolidated interim financial statements should be read in conjunction with the annual consolidated financial statements of Arrow for the year ended December 31, 2004, included in Arrow’s 2004 Annual Report on Form 10-K.

2.   Accumulated Other Comprehensive (Loss) Income (In Thousands)

The following table presents the components, net of tax, of accumulated other comprehensive (loss) income as of September 30, 2005 and December 31, 2004:

3.   Earnings Per Common Share (In Thousands, Except Per Share Amounts)

The following table presents a reconciliation of the numerator and denominator used in the calculation of basic and diluted earnings per common share (EPS) for the three-month and nine-month periods ended September 30, 2005 and 2004:

8

4.   Stock-Based Compensation Plans

As allowed by SFAS No. 123, “Share-Based Payment,” Arrow accounts for its stock-based compensation plans under the recognition and measurement principles of APB Opinion No. 25, “Accounting for Stock Issued to Employees,” and related interpretations.  No stock-based employee compensation cost has been reflected in net income for stock awards granted under these plans (other than for certain stock appreciation rights attached to options granted in 1992 and earlier, all of which have been exercised as of January 2002), as all awards granted under these plans have been options having an exercise price equal to the market value of the underlying common stock on the date of grant.  However, options granted do generally impact diluted earnings per share by increasing the weighted average diluted shares outstanding and thereby decreasing diluted earnings per share as compared to basic earnings per share.

There were no options granted in the first nine months of 2005.  The weighted-average fair value of options granted during 2004 was $8.01 per option.  The fair value of each option grant is estimated on the date of grant using the Black-Scholes option-pricing model with the following weighted-average assumptions used for grants in 2004: dividend yield of 2.88%; expected volatility of 28.4%; risk free interest rate of 3.78%; and expected lives of 7.0 years.  Arrow also sponsors an Employee Stock Purchase Plan (ESPP) under which employees purchased Arrow’s common stock at a 15% discount below market price at the time of purchase for the first two months of 2005 and prior to then.  This discount was changed to 5% discount below market price for all subsequent purchases.  Under APB 25, a plan with a discount of 15% or less is not considered compensatory and expense is not recognized.  Under SFAS No. 123, however, a stock pu rchase plan with a discount in excess of 5% is considered a compensatory plan and thus the ESPP was considered a compensatory plan for the first two months of 2005, and the entire discount for that period was considered compensation expense in the pro forma disclosures set forth below.  The effects of applying SFAS No. 123 on pro forma net income in the recently completed period may not be representative of the effects on pro forma net income for future periods.  

The following table illustrates the effect on net income and earnings per share if we had applied the fair value recognition provisions of SFAS No. 123, “Accounting for Stock-Based Compensation,” to stock-based employee compensation plans.

In December 2004, the FASB issued a revised SFAS No. 123 (SFAS No. 123R), “Share-Based Payment.”  SFAS No. 123R requires that we measure the cost of employee services received in exchange for an award of equity instruments based on the fair value of the award on the grant date.  That cost will be recognized over the period during which an employee is required to provide service in exchange for the award (i.e. the vesting period), which is typically four years for Arrow.  In a press release dated April 14, 2005, the SEC delayed the effective date of SFAS No. 123R to the first quarter of 2006.  

9

5.   Guarantees

Arrow does not issue any guarantees that would require liability-recognition or disclosure, other than its standby letters of credit.  Standby and other letters of credit are conditional commitments issued by Arrow to guarantee the performance of a customer to a third party.  Those guarantees are primarily issued to support public and private borrowing arrangements, including bond financing and similar transactions.  The credit risk involved in issuing letters of credit is essentially the same as that involved in extending loan facilities to customers.  Typically, these instruments have terms of twelve months or less.  Some expire unused, and therefore, the total amounts do not necessarily represent future cash requirements.  Some have automatic renewal provisions.

For letters of credit, the amount of the collateral obtained, if any, is based on management’s credit evaluation of the counter-party.   Arrow had approximately $3.3 million of standby letters of credit on September 30, 2005, most of which will expire within one year and some of which were not collateralized.  At that date, all the letters of credit were for private borrowing arrangements.  The fair value of the Arrow’s standby letters of credit at September 30, 2005 was insignificant.

6.   Retirement Plans (In Thousands)

The following table provides the components of net periodic benefit costs for the three months ended September 30:

The following table provides the components of net periodic benefit costs for the nine months ended September 30:

We previously disclosed in our financial statements for the year ended December 31, 2004 that we do not expect to make a contribution to the qualified defined benefit pension plan during 2005.  However, in the second quarter of 2005 we determined that it was appropriate to contribute approximately $792 thousand to the plan, the maximum actuarially recommended contribution for the 2005 plan year.  

In December 2003, the Medicare Prescription Drug, Improvement and Modernization Act of 2003 (the Act) became law in the United States, however, final regulations were not issued until January 2005.  The Act introduced a prescription drug benefit under Medicare as well as a federal subsidy to sponsors of retiree health care benefit plans that provide a benefit that is at least actuarially equivalent to Medicare Part D under the Act.  The measures of the accumulated non-pension postretirement benefit obligation at September 30, 2005 and net periodic non-pension postretirement benefit cost for the third quarter of 2005 reflect the benefit associated with the subsidy.

10

7.   Recently Issued Accounting Pronouncements

In January 2003, the FASB issued FASB Interpretation No. 46 (“FIN 46”), “Consolidation of Variable Interest Entities”.  The objective of this interpretation was to provide guidance on how to identify a variable interest entity (“VIE”) and determine when the assets, liabilities, non-controlling interests, and results of operations of a VIE need to be included in a company’s consolidated financial statements.  FIN 46 was effective for all VIE’s created after January 31, 2003.  However, the FASB postponed that effective date to December 31, 2003.  In December 2003, the FASB issued a revised FIN 46 (FIN 46 R), which further delayed this effective date until March 31, 2004 for VIE’s created prior to February 1, 2003, except for special purpose entities, which were required to adopt either FIN 46 or FIN 46 R as of December 31, 2003.  The requirements of FIN 46 R resulted in the deconsolidati on of our wholly-owned subsidiary trusts, formed to issue redeemable preferred securities (“trust preferred securities”) to the public, the proceeds of which are used by the trust to acquire subordinated debt of Arrow.  Under final rules issued February 28, 2005 by the Federal Reserve Board, trust preferred securities may continue to qualify as Tier 1 capital for bank regulatory purposes, in an amount not to exceed 25% of Tier 1 capital.  The final rule limits restricted core capital elements to a percentage of the sum of core capital elements, net of goodwill less any associated deferred tax liability.  We have issued trust preferred securities in 2003 and 2004 totaling $20 million.  Up to half of total capital may consist of so-called "Tier 2" capital, comprising a limited amount of subordinated debt, preferred stock not qualifying as Tier 1 capital, certain other instruments and a limited amount of the allowance for loan losses.

Report of Independent Registered Public Accounting Firm

The Board of Directors and Shareholders

Arrow Financial Corporation:

We have reviewed the consolidated balance sheet of Arrow Financial Corporation and subsidiaries (the “Company”) as of September 30, 2005 and the related consolidated statements of income for the three-month and nine month periods ended September 30, 2005 and 2004, and the consolidated statements of changes in shareholders’ equity and cash flows for the nine-month periods ended September 30, 2005 and 2004.  These consolidated financial statements are the responsibility of the Company’s management.

We conducted our reviews in accordance with the standards of the Public Company Accounting Oversight Board (United States).  A review of interim financial information consists principally of applying analytical procedures and making inquiries of persons responsible for financial and accounting matters.  It is substantially less in scope than an audit conducted in accordance with the standards of the Public Company Accounting Oversight Board (United States), the objective of which is the expression of an opinion regarding the financial statements taken as a whole.  Accordingly, we do not express such an opinion.

Based on our review, we are not aware of any material modifications that should be made to the consolidated financial statements referred to above for them to be in conformity with U.S. generally accepted accounting principles.

We have previously audited, in accordance with standards of the Public Company Accounting Oversight Board (United States), the consolidated balance sheet of Arrow Financial Corporation and subsidiaries as of December 31, 2004, and the related consolidated statements of income, changes in shareholders’ equity and cash flows for the year then ended (not presented herein); and in our report dated March 7, 2005, we expressed an unqualified opinion on those consolidated financial statements.  In our opinion, the information set forth in the accompanying consolidated balance sheet as of December 31, 2004, is fairly stated, in all material respects, in relation to the consolidated balance sheet from which it has been derived.

/s/ KPMG LLP

Albany, New York

November 9, 2005

11

Item 2.

ARROW FINANCIAL CORPORATION AND SUBSIDIARIES

MANAGEMENT'S DISCUSSION AND ANALYSIS OF

FINANCIAL CONDITION AND RESULTS OF OPERATIONS

SEPTEMBER 30, 2005

Note on Terminology - In this Quarterly Report on Form 10-Q, the terms “Arrow,” “the registrant,” “we,” “us,” and “our” generally refer to Arrow Financial Corporation and its subsidiaries, as a group, except where the context indicates otherwise.  Arrow is a two-bank holding company headquartered in Glens Falls, New York.  Our banking subsidiaries are Glens Falls National Bank and Trust Company whose main office is located in Glens Falls, New York, and Saratoga National Bank and Trust Company whose main office is located in Saratoga Springs, New York.

At certain points in this Report, our performance is compared with that of our “peer group” of financial institutions.  Peer data has been obtained from the Federal Reserve Board’s June 2005 “Bank Holding Company Performance Report.”  Unless otherwise specifically stated, our peer group is comprised of the group of the 223 domestic bank holding companies identified in that report having from $1 to $3 billion in total consolidated assets.

Forward Looking Statements - The information contained in this Quarterly Report on Form 10-Q contains statements that are not historical in nature but rather are based on our beliefs, assumptions, expectations, estimates and projections about the future.  These statements are “forward-looking statements” within the meaning of Section 21E of the Securities Exchange Act of 1934, as amended, and involve a degree of uncertainty and attendant risk.  Words such as “expects,” “believes,” “anticipates,” “estimates” and variations of such words and similar expressions often identify such forward-looking statements.  Some of these statements, such as those included in the interest rate sensitivity analysis in Item 3, entitled “Quantitative and Qualitative Disclosures About Market Risk,” are merely presentations of what future performance or changes in future performance would look like based on hypothetical assumptions and on simulation models.  Other forward-looking statements are based on our general perceptions of market conditions and trends in activity, both locally and nationally, as well as current management strategies for future operations and development.

Certain forward-looking statements in this Report are referenced in the table below:

These statements are not guarantees of future performance and involve risks and uncertainties that are difficult to quantify or, in some cases, to identify.  In the case of all forward-looking statements, actual outcomes and results may differ materially from what the statements predict or forecast.  

Factors that could cause or contribute to such differences include, but are not limited to, unexpected changes in economic and market conditions, including unanticipated developments in interest rates generally, including changes in the yield curve; sudden changes in consumer spending or consumer confidence; new developments in state and federal regulation; enhanced competition from unforeseen sources; new emerging technologies; unexpected loss of our key personnel; unanticipated market or business opportunities for us; and similar uncertainties inherent in banking operations or business generally.  Readers are cautioned not to place undue reliance on these forward-looking statements, which speak only as of the date hereof.  We undertake no obligation to revise or update these forward-looking statements to reflect the occurrence of unanticipated events.  This Quarterly Report should be read in conjunction with our Annual Report on Form 10- K for December 31, 2004.

12

USE OF NON-GAAP FINANCIAL MEASURES

The Securities and Exchange Commission (SEC) has adopted Regulation G, which applies to all public disclosures, including earnings releases, made by registered companies that contain “non-GAAP financial measures.”  GAAP is generally accepted accounting principles in the United States of America.  Under Regulation G, companies making disclosures containing non-GAAP financial measures must also disclose, along with each non-GAAP financial measure, certain additional information, including a reconciliation of the non-GAAP financial measure to the closest comparable GAAP financial measure and a statement of the company’s reasons for utilizing the non-GAAP financial measure as part of its financial disclosures.  At the same time that the SEC issued Regulation G, it also made amendments to Item 10 of Regulation S-K, requiring companies to make the same types of supplemental disclosures whenever they include non-GAAP financial meas ures in their filings with the SEC.  The SEC has exempted from the definition of “non-GAAP financial measures” certain specific types of commonly used financial measures that are not based on GAAP.  When these exempted measures are included in public disclosures or SEC filings, supplemental information is not required.  The following measures used in this Report which have not been specifically exempted by the SEC may nevertheless constitute "non-GAAP financial measures" within the meaning of the SEC's new rules, although we are unable to state with certainty that the SEC would so regard them.

Tax-Equivalent Net Interest Income and Net Interest Margin: Net interest income, as a component of the tabular presentation by financial institutions of Selected Financial Information regarding their recently completed operations, is commonly presented on a tax-equivalent basis.  That is, to the extent that some component of the institution's net interest income will be exempt from taxation (e.g., was received by the institution as a result of its holdings of state or municipal obligations), an amount equal to the tax benefit derived from that component is added back to the net interest income total.  This adjustment is considered helpful in comparing one financial institution's net interest income (pre-tax) to that of another institution, as each will have a different proportion of tax-exempt items in their portfolios.  Moreover, net interest income is itself a component of a second financial measure commonly use d by financial institutions, net interest margin, which is the ratio of net interest income to average earning assets.  For purposes of this measure as well, tax-equivalent net interest income is generally used by financial institutions, again to provide a better basis of comparison from institution to institution.  We follow these practices.

The Efficiency Ratio: Financial institutions often use an "efficiency ratio" as a measure of expense control.  The efficiency ratio typically is defined as the ratio of noninterest expense to net interest income and noninterest income.  As in the case of net interest income generally, net interest income as utilized in calculating the efficiency ratio is typically expressed on a tax-equivalent basis.  Moreover, most financial institutions, in calculating the efficiency ratio, also adjust both noninterest expense and noninterest income to exclude from these items (as calculated under GAAP) certain component elements, such as intangible asset amortization (deducted from noninterest expense) and securities gains or losses (excluded from noninterest income).  We follow these practices.

13

OVERVIEW

Selected Quarterly Information:

(Dollars In Thousands, Except Per Share Amounts)

Share and Per Share amounts have been restated for the September 2005 3% stock dividend.

14

Selected Quarterly Information, Continued:

¹ See “Use of Non-GAAP Financial Measures” on page 13.

Selected Nine-Month Period Information:

(Dollars In Thousands, Except Per Share Amounts)

Share and Per Share amounts have been restated for the September 2005 3% stock dividend.

15

Selected Nine-Month Period Information, Continued:

¹ See “Use of Non-GAAP Financial Measures” on page 13.

16

Average Consolidated Balance Sheets and Net Interest Income Analysis

(see “Use of Non-GAAP Financial Measures” on page 13)

(Fully Taxable Basis using a marginal tax rate of 35%)

(Dollars In Thousands)

17

Average Consolidated Balance Sheets and Net Interest Income Analysis

(see “Use of Non-GAAP Financial Measures” on page 13)

(Fully Taxable Basis using a marginal tax rate of 35%)

(Dollars In Thousands)

18

OVERVIEW

We reported earnings of $4.839 million for the third quarter of 2005, a decrease of $129 thousand, or 2.6%, as compared to $4.968 million for the third quarter of 2004.  Diluted earnings per share were $.46 and $.47 for the respective quarters.  Average shares outstanding decreased between the two periods, representing the effect of our common stock repurchases, offset in part by issuances of our common stock under our compensatory stock plans.  For the first nine months of 2005 we reported earnings of $13.949 million, a decrease of $582 thousand, or 4.0%, as compared to $14.531 million for the first nine months of 2004.  Diluted earnings per share were $1.31 and $1.36 for the respective 2005 and 2004 nine-month periods.

The return on average assets for the third quarter of 2005 was 1.31%, compared to 1.42% for the third quarter of 2004, a decrease of 11 basis points, or 7.7%.  The return on average equity for the third quarter of 2005 was 16.39%, compared to 17.87% for the third quarter of 2004, a decrease of 148 basis points, or 8.3%.  For the first nine months of 2005, the return on average assets was 1.30%, compared to 1.40% for the prior year period, a decrease of 10 basis points, or 7.1%.  The return on average equity for the first nine months of 2005 was 15.90%, compared to 17.70% for the prior year period, a decrease of 180 basis points, or 10.2%.

The principal reason for the declining earnings ratios from the 2004 periods to the 2005 periods was the decrease in net interest margin and net interest income.  This development, in turn, reflected the fact that our cost of funds, principally interest-paying deposits, increased more rapidly than the average yield on our assets, principally our loan portfolio.  Our experience in this respect, of diminishing margins, mirrored that of financial institutions generally.  For the third quarter of 2005, net interest margin was 3.63%, down 33 basis points, or 8.3%, from the 3.96% margin for the third quarter of 2004 and down 9 basis points, or 2.4%, from the second quarter of 2005.

To a limited extent, the negative impact of our declining margins in recent periods has been offset by growth in our asset base.  Total assets were $1.48 billion at September 30, 2005, which represented an increase of $106.2 million, or 7.7%, from December 31, 2004, and an increase of $99.3 million, or 7.2%, above the level at September 30, 2004.  In each case, the increase was impacted by the acquisition of three branches in the second quarter of 2005, which is discussed on the next page.

Total shareholders’ equity was $116.0 million at September 30, 2005, a decrease of $2.0 million, or 1.7%, from December 31, 2004.  The increase from retained earnings and stock issuances under our share plans was more than offset by the decrease in our equity resulting from repurchases of our common stock and unrealized losses in the available-for-sale securities portfolio.  Our risk-based capital ratios and Tier 1 leverage ratio continued to exceed regulatory minimum requirements at period-end.  At September 30, 2005 both our banks qualified as "well-capitalized" under federal bank guidelines.  Efficient utilization of capital remains a high priority of Arrow.

CHANGE IN FINANCIAL CONDITION

Summary of Selected Consolidated Balance Sheet Data

(Dollars in Thousands)

(1) Includes Nonaccrual Loans

19

Branch Acquisition:  Our acquisition of three HSBC branches on April 8, 2005 included approximately $62 million in deposit balances and $8 million in loan balances.  The acquisition also resulted in an increase of $5.9 million in intangible assets.  Deposit balances at the acquired branches through September 30, 2005 generally stayed level.  Changes to our balance sheet, described in the next two paragraphs, include the impact of the branch acquisition.

Sources of Funds:  Our deposit balances increased $106.7 million over the first nine months of 2005 and accounted for nearly all of the increase in our sources of funds.  The combined balance of short-term borrowings and Federal Home Loan Bank (FHLB) advances decreased slightly.  Municipal deposits are a significant segment of our deposit balances and represented 16.7% of total deposits at September 30, 2005.  Municipal deposits, which typically are at seasonal lows during the summer, increased by $19.2 million from June 30, 2005 to September 30, 2005, but were essentially unchanged from the 2004 year-end balance.  Consumer and business deposit balances increased by $107.8 million from December 31, 2004 to September 30, 2005.  During the first nine months of 2005, federal funds sold and investment securities were actually a source of funds, as the combination of sales, maturities and amortization of mortgage-ba cked securities offset purchases of new securities by approximately $14.0 million.

Deployment of Funds into Earning Assets:  The primary use of our increased sources of funds, cited above, from December 31, 2004 to September 30, 2005 was to fund a $106.0 net increase in the loan portfolio.  Over half of the increase ($59.6 million) was in our largest loan segment, indirect consumer loans (principally automobile financing).  However, we also experienced increases in all our other loan segments: commercial, residential real estate and other consumer lending.  The fact that total assets increased $106.2 million over the nine-month period, while earning assets only increased $85.0 million was primarily attributable to two items: i) the balance in cash and due from banks was unusually low at December 31, 2004, at $29.8 million compared to the 2004 year-to-date average balance of $35.9 million and $41.4 million at period-end September 30, 2005, and ii) an increase in intangible assets of $5.6 million (primar ily due to the branch acquisition).

Deposit Trends

The following two tables provide information on trends in the balance and mix of our deposit portfolio by presenting, for each of the last five quarters, the quarterly average balances by deposit type and the percentage of total deposits represented by each deposit type.

Quarterly Average Deposit Balances

(Dollars in Thousands)

Percentage of Average Quarterly Deposits

For a variety of reasons, including the seasonality of municipal deposits noted above, we typically experience little net deposit growth in the first quarter of the year, but more significant growth in the second quarter.  Average deposit balances followed this pattern for the first two quarters of 2005.  The average balance of deposits actually decreased from the fourth quarter of 2004 to the first quarter of 2005.  Excluding the impact of the branch acquisition, all categories of deposits (except regular and money market savings) experienced growth in the second quarter of 2005 above first quarter balances.  Although average deposit balances for the third quarter of 2005 (when municipal deposits are at a seasonal low) were down slightly from the second quarter average balances we did experience increases in consumer and business account balances.

20

During the uninterrupted period of declining interest rates from May 2000 through the first half of 2004, we experienced a trend (typical for financial institutions) of maturing time deposits transferring to non-maturity transaction accounts.  This period of declining rates ended in June 2004 as the Federal Reserve initiated a series of twelve 25 basis point increases in prevailing short-term rates (federal funds rate) extending through early November 2005.  As a result of this rising short-term rate environment we began to experience a reversal of the prior trend in deposit account migration as our customers, including municipal accounts, started to transfer some of their non-maturity balances back into high-yielding time deposits (although with maturities of fairly short duration).  At September 30, 2005 time deposits represented 29.1% of total deposits, up from 22.7% at June 30, 2004.  This higher percentage was still well below th e ratio of 41.0% at June 30, 2000.

In addition to the three branches acquired in April 2005, we opened a new branch in Queensbury, New York in June 2004.  Otherwise, the increase in deposits between the two periods was achieved through our existing base of branches.  We have no brokered deposits.

Quarterly Average Rate Paid on Deposits

Key Interest Rate Changes 1999 – 2005

Our net interest income has traditionally been sensitive to and impacted by changes in prevailing market interest rates, due to the fact that our deposit liabilities tend to reprice more rapidly than our earning assets.  Thus generally there has been a negative correlation between changes in interest rates (particularly during periods when interest rates begin to change direction), and our net interest income in immediately ensuing periods.  When interest rates begin to decline, net interest income has increased in ensuing periods, and vice versa.  

21

In the first quarter of 2001, after an extended period of stable interest rates and six months of moderate rate increases, the Federal Reserve Board reversed direction and began decreasing short-term interest rates rapidly and significantly in response to perceived weakening in the economy.  By December 2001, the total decrease in prevailing short-term interest rates for the year was 425 basis points.  In the first eleven months of 2002, there were no rate changes, but the Federal Reserve Board then decreased rates another 50 basis points in November 2002.  As a result of this rapid and substantial rate decrease, we experienced a decrease in the cost of our deposits not only in 2001, but for the next three years as well, although the Federal Reserve did not reduce short-term rates again until June 2003.  Yet although our deposit rates began to decrease in the first quarter of 2001, we did not experience a decrease in the average yield in our loan portfolio until the second quarter of 2001.  Yields on our loan portfolio continued to decrease through the remainder of 2001 and for the next three years, and into 2005.  See the “Loan Trends” section in this Report beginning on page 23, for a more complete analysis of yield trends in the loan portfolio.  The net effect of the Federal Reserve’s rate decreases commencing in 2001 was, at least initially, very positive for our net interest margin and net interest income.

During 2003 and the first half of 2004, the effect of the Federal Reserve’s rate decreases on our deposit rates began to diminish, because rates on several of our deposit products, such as savings and NOW accounts, were already priced at such low levels that further significant decreases in the rates for such products was not practical or sustainable.  Yields on our loan portfolio, however, continued to fall significantly in 2003 and 2004, putting serious pressure on the net interest margin.  At this point, the impact of the decreasing rate environment turned negative, as yields on earning assets decreased more rapidly than rates paid on paying liabilities, and net interest margin began to shrink.

The net interest margin for the full year of 2003 was 4.05%, a decrease of 45 basis points, or 10.0%, from the prior year.  During 2003 the yields on earning assets fell 91 basis points, while the cost of paying liabilities fell only 57 basis points.  In the first quarter of 2004 the net interest margin stabilized briefly at 3.97%, an increase of 2 basis points from the net interest margin for the fourth quarter of 2003, as both the yields on average earning assets and the cost of paying liabilities decreased slightly from the fourth quarter of 2003.  However, the narrowing of the net interest margin resumed in the second quarter of 2004 as it decreased 11 basis points from the net interest margin for the first quarter of 2004.  Yields on average earning assets decreased 14 basis points from the first quarter, while the cost of paying liabilities decreased only 4 basis points.

In the second quarter of 2004, the Federal Reserve reversed direction and began to increase prevailing rates with five successive 25 basis point increases in the federal funds rate in the remainder of 2004.  This change in direction did not immediately impact either our cost of paying liabilities or our yield on earning assets, both because of normal time-lag in the responsiveness of our rates to Federal Reserve actions and for reasons unique to our portfolios.  The change in the mix of our total deposits in the third quarter of 2004, reflecting our decision to de-emphasize certain high cost municipal deposits, resulted in a reduced average rate paid on deposits during the quarter, with a resulting positive impact on net interest margin, which increased by 10 basis points to 3.96% from the second quarter.  The cost of all NOW accounts fell from 1.21% for the second quarter to .80% for the third quarter.  The yield on earning assets, m eanwhile, only decreased 3 basis points from the prior quarter.  By the fourth quarter of 2004, the increases in the target federal funds rate started to have the expected impact on our net interest margin.  Rates paid on new time deposits began to exceed the rates on maturing time deposits.  Thus, our net interest margin for the fourth quarter of 2004 decreased by 5 basis points from the third quarter to 3.91%.  Meanwhile, there was very little change in rates earned on average earning assets from the third quarter to the fourth quarter.  As expected in periods where interest rates begin to rise, our net interest margin continued to decrease, as in the aggregate, the cost of deposits and other borrowed funds was increasing faster than rates on earning assets.

During the first three quarters of 2005, this trend of tightening margins continued.  The Federal Reserve made six more 25 basis point increases in prevailing rates.  While we resisted increasing rates on our non-maturity deposit products, we did increase rates paid on time deposits.   Maturing time deposits began to reprice at higher rates. As noted above, we also began to experience a significant shift from non-maturity deposits to time deposits.  The costs of total deposits increased 11 basis points during the first quarter, to 1.19% from 1.08% from the fourth quarter of 2004, 15 basis points in the second quarter of 2005 to 1.34% and another 11 basis points in the third quarter of 2005 to 1.45%.  Meanwhile, the taxable equivalent yield on our loan portfolio lagged behind our deposit rate movement; yields decreased 3 basis points from the fourth quarter of 2004 to the first quarter of 2005, stayed flat at 5.76% for the se cond quarter of 2005 and only increased another 3 basis points to 5.79% during the third quarter of 2005.

In both rising and falling rate environments, we face significant competitive pricing pressures in our marketplace for both deposits and loans, and thus ultimately both assets and liabilities may be expected to reprice proportionately in response to changes in market rates.

22

Non-Deposit Sources of Funds

We have borrowed funds from the Federal Home Loan Bank ("FHLB") under a variety of programs, including fixed and variable rate short-term borrowings and borrowings in the form of "convertible advances."  These convertible advances have original maturities of 2 to 10 years and are callable by the FHLB at certain dates beginning no earlier than one year from the issuance date.  If the advances are called, we may elect to receive replacement advances from the FHLB at the then prevailing FHLB rates of interest.  At September 30, 2005 our FHLB borrowings included $31.5 million of overnight borrowings and term advances of $100.0 million.

In each of 2004 and 2003 we privately placed $10 million of capital securities issued by subsidiary Delaware business trusts specifically formed for the purpose of facilitating such funding.  The proceeds of the offering of these “trust preferred securities” were invested in a like amount of subordinated debentures issued by our holding company.  The securities are reflected as “Junior Subordinated Obligations Issued to Unconsolidated Subsidiary Trusts” on our consolidated balance sheet as of September 30, 2005.  The securities have certain features that make them an attractive funding vehicle, principally their status as qualifying regulatory capital.  Under final rules issued February 28, 2005 by the Federal Reserve, trust preferred securities may qualify as Tier 1 capital, in an amount not to exceed 25% of total Tier 1 capital, net of goodwill less any associated deferred tax liability.  Both of our trust p referred issues qualify as regulatory capital under capital adequacy guidelines discussed below.  However, both issues of trust preferred securities are subject to early redemption by us if the proceeds cease to qualify as Tier 1 capital of Arrow, which would only happen if bank regulatory authorities were to reverse their current position that trust preferred securities issued by subsidiaries of bank holding companies, up to certain threshold levels, qualify for such treatment.

Loan Trends

The following two tables present, for each of the last five quarters, the quarterly average balances by loan type and the percentage of total loans represented by each loan type.

Quarterly Average Loan Balances

(Dollars in Thousands)

Percentage of Quarterly Average Loans

¹ Other Consumer Loans includes certain home improvement loans, secured by mortgages, in this table of average loan balances.

23

Indirect Loans: For several years preceding the third quarter of 2001, the indirect consumer loan portfolio (consisting principally of auto loans financed through local dealerships where we acquire the dealer paper) was the fastest growing segment of our loan portfolio, both in terms of absolute dollar amount and as a percentage of the overall portfolio.  Over the subsequent quarters, this segment of the portfolio, while remaining the largest in total outstanding balances, ceased to grow in absolute terms and decreased as a percentage of the overall portfolio.  This flattening of indirect loan totals was largely the result of aggressive campaigns of zero rate and other subsidized financing by auto manufacturers, commencing in the fall of 2001.  During the fourth quarter of 2002, and for the first two quarters of 2003, the indirect portfolio experienced a small amount of growth as we became more rate competitive, but the level of indirect loans was flat for the third quarter of 2003 and decreased by $11.9 million during the fourth quarter of 2003.  During the first half of 2004 indirect loan balances continued to decline, and then rose slightly during the second half of the year.  

At the end of the first quarter of 2005, we experienced an increase in indirect loans, which did not have a large impact on the average balance for the quarter (an $841 thousand increase from the prior quarter), but did cause the balance at period-end to rise sharply to $312.9 million.  We continued to experience strong demand for indirect loans throughout the second and third quarters of 2005, for a variety of factors, including modifications by the automobile manufacturers of their subsidized financing programs.  The average balances increased by $21.7 million, or 7.1%, from the first quarter to the second quarter and by another $29.8 million, or 9.1%, in the third quarter.

Indirect loans still represent the largest category of loans (36.0%) in our portfolio, and any developments threatening our indirect loan business generally may be expected to have a negative impact on our financial performance.  If auto manufacturers resume their heavily subsidized financing programs, our indirect loan portfolio is likely to continue to experience rate pressure and limited, if any, overall growth.

Residential Real Estate Loans: Residential real estate loans represented the second largest segment of our loan portfolio at September 30, 2005, at 30.2% of average loans for the quarter.  This segment of our portfolio increased by $14.3 million from year-end 2003 to 2004, despite the fact that we sold (with servicing retained) $15.4 million of fixed low-rate mortgages during 2004.  To date in 2005, we have retained all but $5.9 million of the $52.8 million of residential real estate loans originated.  In the first quarter of 2005, these originations were not able to keep up with principal amortization and payoffs resulting in a net decrease in the average balance from the fourth quarter of 2004.  However, demand increased during the second and third quarters of 2005.  In the second quarter, the average balance increased by $5.9 million, or 2.1%, and in the third quarter, it increased another $5.7 million, or 2.0%.  Residential mortgage demand has remained moderate during 2004 and into 2005 and we expect that, if we continue to retain all or most originations, we will be able to maintain the level of residential real estate loans and may experience some continued growth.

Commercial and Commercial Real Estate Loans: We have experienced strong to moderate demand for commercial loans in recent periods, a trend that has persisted for several years.  Commercial and commercial real estate loans have grown each year for the past five years, both in dollar amount and as a percentage of the overall loan portfolio.  These loans represented the fastest growing segment in our loan portfolio for the first quarter of 2005.  In the second quarter, the average balance of commercial and commercial real estate loans increased by $9.2 million, or 4.4%, and in the third quarter it increased by another $3.8 million, or 1.8%.

Quarterly Taxable Equivalent Yield on Loans

24

In general, the yield (tax-equivalent interest income divided by average loans) on our loan portfolio and other earning assets has been impacted by changes in prevailing interest rates, as previously discussed on pages 21 and 22 under the heading "Key Interest Rate Changes 1999 - 2005."  We expect that such will continue to be the case, that is, that loan yields will continue to rise and fall with changes in prevailing market rates, although the timing and degree of responsiveness will continue to be influenced by a variety of other factors, including the makeup of the loan portfolio, the yield curve (i.e. the relationship between short-term rates and long-term rates), consumer expectations and preferences, competition and the rate at which the portfolio expands.  Many of the loans in the commercial portfolio have variable rates tied to prime, FHLB or U.S. Treasury indices, although in recent periods these indices, typically reflectiv e of long-term rates, have not changed as rapidly or as significantly as short-term rates.   There is a significant amount of cash flow from normal amortization and prepayments in all loan categories, and the portfolio reprices at current rates as new loans are generated at the current yields or existing loans are renegotiated.  As noted in the earlier discussion, during the recently concluded long period of declining rates (from early 2001 to mid-2004), we experienced a time lag between the impact of declining rates on the deposit portfolio (which was felt relatively quickly) and the impact on the loan portfolio (which occurred more slowly).  During early stages of this decline, the time lag had a positive impact on net interest margin, followed by a negative impact in the later states of the decline.

The net interest margin expanded during 2001 and into the first quarter of 2002 as deposit rates decreased rapidly.  Our deposit rates began to flatten out in mid-2002, while loan yields continued to decline.  As a result, the net interest margin began to contract in the second quarter of 2002.  Generally, this pattern persisted through the remainder of 2002, all of 2003 and through the third quarter of 2004, with the cost of deposits decreasing slightly, and loan yields decreasing somewhat faster.  By the fourth quarter of 2004, the cost of our deposits actually began to increase while the yields on loans were still slowly decreasing, as the yield on new loan originations generally was still below the portfolio average.  The result was continuing shrinkage of the net interest margin.  The average yield on the loan portfolio decreased 1 basis point in the fourth quarter of 2004, 3 basis points in the first quarter of 2005, s tayed flat for the second quarter of 2005 and only increased by 3 basis points in the third quarter of 2005.  This put increasing pressure not only on net interest margin, which diminished significantly over the first three quarters of 2005, but also on net interest income, which also decreased over the period, despite continued asset growth.  During the first quarter of 2005, we experienced a significant decrease in net interest income in comparison to the prior-year quarter.  For the second quarter of 2005, net interest income was virtually the same as the prior-year second quarter, but net interest income for the third quarter of 2005 was approximately 3% below the third quarter of 2004.

If rates, especially long-term rates, continue to increase in forthcoming periods, as many analysts have forecasted, we expect that ultimately the yield on our loan portfolio will start to increase, matching the increasing cost of deposits that we are already experiencing.  We can give no assurances that this will happen, however.  Unless and until it does, we may experience further reductions in our net interest margin, to the detriment of net interest income and earnings generally.

25

Asset Quality

The following table presents information related to our allowance and provision for loan losses for the past five quarters.  

Summary of the Allowance and Provision for Loan Losses

(Dollars in Thousands)(Loans Stated Net of Unearned Income)

26

Our nonperforming assets at September 30, 2005 amounted to $2.3 million, an increase of $70 thousand, or 3.1%, from December 31, 2004, and a decrease of $958 thousand, or 29.3%, from September 30, 2004.  

At period-end, nonperforming assets represented .16% of total assets, unchanged from year-end 2004 and an 8 basis point decrease from .24% at September 30, 2004.  The current ratio is at or near our historical low.  At June 30, 2005 the ratio of nonperforming assets to total assets for our peer group was .50%.

The balance of other non-current loans as to which interest income was being accrued (i.e. loans 30-89 days past due as defined in bank regulatory guidelines) totaled $4.5 million at September 30, 2005 and represented 0.46% of loans outstanding at that date, as compared to approximately $4.6 million of non-current, but accruing loans at December 31, 2004 representing 0.52% of loans outstanding.  These non-current, but accruing loans at September 30, 2005 were composed of approximately $3.6 million of consumer loans, principally indirect automobile loans, $.7 million of residential real estate loans and commercial loans of $.2 million.

The percentage of our performing loans that demonstrate characteristics of potential weakness from time to time, typically a very small percentage, depend principally on economic conditions in our geographic market area of northeastern New York State.  In general, the economy in this area has been relatively strong in recent periods, extending back two or three years.  The regional economy was healthy from 1997-2000, and when the U.S. experienced a mild recession in 2001 and 2002, the economic downturn was not as severe in our geographic market area as in most areas.  During that periods and in ensuing years, the unemployment rate in the "Capital District" in and around Albany and areas north, including our principal service areas in Warren and Washington counties, has been at or below the national average.  However, in our other service areas including Clinton and Essex Counties, near the Canadian border, the unemployment r ate has been at or slightly above the national average in recent months.

The ratio of the 2005 third quarter net charge-offs to average loans (annualized) was .07%, up 1 basis point from the ratio for the third quarter of 2004.  The provision for loan losses was $218 thousand for the third quarter of 2005, compared to a provision of $205 thousand for the third quarter of 2004.  The provision as a percentage of average loans (annualized) was .09% for the third quarter of 2005, unchanged from the ratio for the comparable 2004 period.  

The allowance for loan losses at September 30, 2005 amounted to $12.2 million, or 1.24% of outstanding loans, 14 basis points lower than the ratio at December 31, 2004 and also 13 basis points lower than the ratio at September 30, 2004.   The allowance as a percent of nonperforming loans was 588.83% at September 30, 2005.

27

CAPITAL RESOURCES

Shareholders' equity decreased $2.0 million during the first nine months of 2005.  During the period, net income of $13.9 million was reduced by stock repurchases (net of new stock issuances through stock plans) totaling $5.9 million, cash dividends of $7.1 million ($.68 per share) and net unrealized losses on securities available-for-sale (net of tax) of approximately $3.7 million. From September 30, 2004 to September 30, 2005, shareholders' equity increased by $2.9 million, or 2.5%.  Current and prior period changes in shareholders' equity are presented in the Consolidated Statements of Changes in Shareholders' Equity, on pages 5 and 6 of this report.

On April 27, 2005 the Board of Directors approved a stock repurchase program authorizing the repurchase, at the discretion of senior management, of up to $5 million of Arrow’s common stock over the next twelve months in open market or negotiated transactions.  This program replaced a similar stock repurchase program, approved by the Board in April 2004, under which we repurchased approximately $3.7 million of common stock out of the $5 million authorized for repurchase.  See Part II, Item 2 of this Report for further information on stock repurchases and repurchase programs.

The following discussion of capital focuses on regulatory capital ratios, as defined and mandated for financial institutions by federal bank regulatory authorities.  Regulatory capital, although a financial measure that is not provided for or governed by GAAP, nevertheless has been exempted by the SEC from the definition of "non-GAAP financial measures" in the SEC's Regulation G governing disclosure of non-GAAP financial measures.  (See the note on page 13 regarding Non-GAAP Financial Measures.)  Thus, certain information which is required to be presented in connection with disclosure of non-GAAP financial measures need not be provided, and has not been provided, for the regulatory capital measures discussed below.  

Our holding company and both of our subsidiary banks are currently subject to two sets of regulatory capital measures, a leverage ratio test and risk-based capital guidelines.  The risk-based guidelines assign risk weightings to all assets and certain off-balance sheet items of financial institutions and establish an 8% minimum ratio of qualified total capital to risk-weighted assets.  At least half of total capital must consist of "Tier 1" capital, which comprises common equity and common equity equivalents, retained earnings, a limited amount of permanent preferred stock and a limited amount of trust preferred securities, less intangible assets.  Up to half of total capital may consist of so-called "Tier 2" capital, comprising a limited amount of subordinated debt, other preferred stock, certain other instruments and a limited amount of the allowance for loan losses.  The third regulatory capital measure, the lev erage ratio test, establishes minimum limits on the ratio of Tier 1 capital to total tangible assets, without risk weighting.  For top-rated companies, the minimum leverage ratio is 3%, but lower-rated or rapidly expanding companies may be required to meet substantially higher minimum leverage ratios.  Federal banking law mandates certain actions to be taken by banking regulators for financial institutions that are deemed undercapitalized as measured by these ratios.  The law establishes five levels of capitalization for financial institutions ranging from "well-capitalized" to "critically undercapitalized."  Federal banking law ties the ability of banking organizations to engage in certain types of non-banking financial activities to such organizations' continuing to qualify as "well-capitalized" under these standards.  

In each of 2003 and 2004, we issued $10 million of trust preferred securities in a private placement.  (See the discussion of trust preferred securities on page 23 under “Non-Deposit Sources of Funds.”)  Under final rules issued by the Federal Reserve Board on February 28, 2005, these securities may qualify as Tier 1 capital in an amount not to exceed 25% of Tier 1 capital, net of goodwill less any associated deferred tax liability.

As of September 30, 2005, the Tier 1 leverage and risk-based capital ratios for Arrow and our subsidiary banks were as follows:  

Summary of Capital Ratios