Changes

Jump to navigation Jump to search
Update table to cover additional shares from cash, other tweaks.
Line 32: Line 32:  
The choice of which class of ETF shares to use, distributing or accumulating, makes no difference to the long-term results generated by the fund. There are some cash flow and possible tax effects, but for two ETFs containing the exact same assets, one distributing and the other accumulating, the investment returns are the same.
 
The choice of which class of ETF shares to use, distributing or accumulating, makes no difference to the long-term results generated by the fund. There are some cash flow and possible tax effects, but for two ETFs containing the exact same assets, one distributing and the other accumulating, the investment returns are the same.
   −
Here is an example. Suppose two ETFs, one named DIST that is distributing, and one named ACCU that is accumulating. Both are launched on the same day with an initial NAV of $1, and both contain the exact same assets. The assets return 10% annually, 7% as capital gains and 3% as dividends. ACCU reinvests the 3% dividend internally into more of the same assets that it already holds, DIST pays this 3% to investors as a dividend distribution.
+
Here is an example. Suppose two ETFs, one named DIST that is distributing, and one named ACCM that is accumulating. Both are launched on the same day with an initial NAV of $1, and both contain the exact same assets. The assets return 10% annually, 7% as capital gains and 3% as dividends. ACCM reinvests the 3% dividend internally into more of the same assets that it already holds, DIST pays this 3% to investors as a dividend distribution.
   −
The following table shows the relative results of two investors, one who buys $10,000 of DIST and the other who buys $10,000 of ACCU at launch. The investor holding DIST reinvests their entire annual dividend distribution into DIST shares as they receive them, and the investor holding ACCU does nothing. The table ignores both tax and trading costs, and assumes no capital gains or other non-dividend distributions.
+
The following table shows the relative results of two investors, one who buys $10,000 of DIST and the other who buys $10,000 of ACCM at launch. The investor holding DIST reinvests their entire annual dividend distribution into DIST shares as they receive it, and the investor holding ACCM does nothing. The table ignores both tax and trading costs, and assumes no capital gains or other non-dividend distributions.
    
{| class="wikitable"
 
{| class="wikitable"
Line 44: Line 44:  
|-
 
|-
 
! scope="col" | Year
 
! scope="col" | Year
! scope="col" | ACCU net asset value
+
! scope="col" | ACCM net asset value
 
! scope="col" | Shares held
 
! scope="col" | Shares held
 
! scope="col" | Holding balance
 
! scope="col" | Holding balance
Line 60: Line 60:  
| 4 || $1.3310 || 10000 || $13,310.00 || $1.2250 || 10864 || $13,308.87 || $1.02
 
| 4 || $1.3310 || 10000 || $13,310.00 || $1.2250 || 10864 || $13,308.87 || $1.02
 
|-
 
|-
| 5 || $1.4641 || 10000 || $14,641.00 || $1.3108 || 11168 || $14,638.97 || $1.81
+
| 5 || $1.4641 || 10000 || $14,641.00 || $1.3108 || 11169 || $14,640.28 || $0.50
 
|-
 
|-
| 6 || $1.6105 || 10000 || $16,105.10 || $1.4026 || 11481 || $16,102.70 || $1.98
+
| 6 || $1.6105 || 10000 || $16,105.10 || $1.4026 || 11482 || $16,104.10 || $0.71
 
|-
 
|-
| 7 || $1.7716 || 10000 || $17,715.61 || $1.5007 || 11802 || $17,711.62 || $3.32
+
| 7 || $1.7716 || 10000 || $17,715.61 || $1.5007 || 11804 || $17,714.62 || $0.59
 
|-
 
|-
| 8 || $1.9487 || 10000 || $19,487.17 || $1.6058 || 12132 || $19,481.34 || $4.76
+
| 8 || $1.9487 || 10000 || $19,487.17 || $1.6058 || 12135 || $19,486.16 || $0.52
 
|-
 
|-
| 9 || $2.1436 || 10000 || $21,435.89 || $1.7182 || 12472 || $21,429.22 || $5.02
+
| 9 || $2.1436 || 10000 || $21,435.89 || $1.7182 || 12475 || $21,434.37 || $0.92
 
|-
 
|-
| 10 || $2.3579 || 10000 || $23,579.48 || $1.8385 || 12821 || $23,570.89 || $6.28
+
| 10 || $2.3579 || 10000 || $23,579.48 || $1.8385 || 12825 || $23,578.24 || $0.49
 
|}
 
|}
   −
At the start of year 1, the both investors hold the same number of shares. At the end of the year, ACCU's net asset value is $1.10, because the ETF has internally reinvested the 3% dividend paid out by the stocks it holds into more of the same stocks. DIST's net asset value is $1.07. However, the investor holding DIST uses the 3% dividend distribution, $300, to purchase 280 more shares of DIST, and leaving them with $0.40 in residual cash because shares can only be purchased and held in whole numbers. The final position of both investors is the same.
+
At the start of year 1, the both investors hold the same number of shares. At the end of the year, ACCM's net asset value is $1.10, because the ETF has internally reinvested the 3% dividend paid out by the stocks it holds into more of the same stocks. DIST's net asset value is $1.07. However, the investor holding DIST uses the 3% dividend distribution, $300, to purchase 280 more shares of DIST, and leaving them with $0.40 in residual cash because shares can only be purchased and held in whole numbers. The final position of both investors is the same.
   −
At the end of year 2, the same happens. ACCU's net asset value rises to $1.21, DIST's to $1.1449, and the investor holding DIST uses the 3% dividend distribution, $329.99, to purchase 288 more shares of DIST at $1.1449/share, again leaving a tiny amount of residual cash but with the two investors otherwise in the same final position.
+
At the end of year 2, the same happens. ACCM's net asset value rises to $1.21, DIST's to $1.1449, and the investor holding DIST uses the 3% dividend distribution, $329.99, to purchase 288 more shares of DIST at $1.1449/share, again leaving a tiny amount of residual cash but with the two investors otherwise in the same final position.
   −
At the end of ten years, although the investor holding DIST shares has suffered a tiny amount of cash drag because they cannot hold fractional ETF shares, the results are the same to within 99.99%. If fractional shares were available, the results would be completely identical.
+
This repeats for years 3 to 10. In years 5, 7, 8 and 10, enough residual cash has accumulated to allow the investor holding DIST to buy one more share than is covered by the dividend paid out to them by the ETF.
   −
Although the example uses USD, this result is completely general to ETFs denominated or traded in any currency.
+
At the end of ten years, although the investor holding DIST shares has suffered a tiny amount of cash drag because they cannot hold fractional ETF shares, the results are the same to within 99.997%. If fractional shares were available, the results would be exactly identical.
 +
 
 +
Although the example uses USD, this result is general to ETFs denominated or traded in any currency.
    
==Effect of taxes==
 
==Effect of taxes==
Line 94: Line 96:     
In practice, when moving from 'accumulation' to 'decumulation' it may be difficult to switch between the two types of ETF without incurring a significant capital gains tax liability, at least outside of tax-sheltered accounts. However, the effect of trading costs will generally be relatively low compared to the likely tax costs, so for most investors the tax effects will dominate.
 
In practice, when moving from 'accumulation' to 'decumulation' it may be difficult to switch between the two types of ETF without incurring a significant capital gains tax liability, at least outside of tax-sheltered accounts. However, the effect of trading costs will generally be relatively low compared to the likely tax costs, so for most investors the tax effects will dominate.
 +
 +
==Effect of large NAVs==
 +
 +
A distributing ETF with a large NAV can result larger cash drag, due to inability of investors to hold fractional shares, particularly for relatively small purchases when reinvesting dividends. Using an equivalent accumulating ETF, if one is available, will help to mitigate this problem.
    
==Comparing real distributing and accumulating ETFs==
 
==Comparing real distributing and accumulating ETFs==
Line 101: Line 107:  
If 'total return' charts or statistics are available for each ETF then this should provide the best comparison. However, even here some care is required. For example, a 'total return' chart might assume a particular tax rate on dividends that is at best an estimate and which will distort the comparison.
 
If 'total return' charts or statistics are available for each ETF then this should provide the best comparison. However, even here some care is required. For example, a 'total return' chart might assume a particular tax rate on dividends that is at best an estimate and which will distort the comparison.
   −
Overall though, it should be safe to assume that apart from differences in tracking error, two ETFs that track the exact same index, one distributing and the other accumulating, should produce broadly identical results, because the assets they hold will themselves produced broadly identical results.
+
Overall though, it should be safe to assume that apart from differences in tracking error, two ETFs that track the exact same index, one distributing and the other accumulating, should produce broadly identical results, because the assets they hold will themselves produced broadly identical results. While local tax laws may change the final outcome, the fact that one ETF is accumulating and the other distributing does not affect the results returned by the ETF by the investor.
    
==See also==
 
==See also==
html_editors
1,794

edits

Navigation menu