Blueprint
|
Simpson Thacher & Bartlett llp
|
|
425 lexington avenue
new york, ny 10017-3954
|
telephone: +1-212-455-2000
facsimile: +1-212-455-2502
|
|
Direct
Dial Number
(212) 455-3066
|
|
E-mail
Address
jmercado@stblaw.com
|
DRAFT
RESPONSE LETTER FOR DISCUSSION WITH SEC STAFF
May 15,
2019
Eric
McPhee
Office
of Real Estate and Commodities
Securities
and Exchange Commission
Division
of Corporation Finance
100 F
Street, NE
Washington,
D.C. 20549
Re:
IRSA Inversiones y Representaciones Sociedad
Anónima
Form 20-F for the fiscal year ended June
30, 2018
Filed October 23, 2018
File
No. 000-30982
Dear
Mr. Telewicz:
On
behalf of our client IRSA Inversiones y Representaciones
Sociedad Anónima (the
“Company”), we are writing to respond to comments
raised in the Staff’s comment letter dated March 25, 2019
(the “Comment Letter”) relating to the above-referenced
annual report (the “Annual Report”) of the Company
filed on October 23,
2018, pursuant to the
Securities Exchange Act of 1934, as amended.
We are
providing the following responses to the comments communicated by
the Staff in the Comment Letter. For convenient reference, we have
reproduced below in bold the text of the comments from the Staff.
The responses and information described below are based upon
information provided to us by the Company.
Form 20-F for the fiscal year ended June 30, 2018
Item 5. Operating and Financial Review and Prospects
Effects of foreign currency fluctuation, page 115
1. We understand that there have been no observable comparable
shopping mall transactions in a number of years, that your shopping
mall valuation methodology relies on a number of inputs that are
not observable or readily estimable (e.g., long-term Argentine
inflation, peso discount rates and peso-dollar forward foreign
exchange rates), and that dislocations in the Argentine economy
affect your valuation approach; please explain:
●
How
you considered the reliability of your shopping mall fair value
measurements and IAS 40.31 when changing from cost to fair
value
●
How
you determined that shopping mall fair value are reliably
measurable on a continuing basis as contemplated in IAS
40.53
In response to the Staff’s comment, the Company respectfully
advises the Staff that IAS
40.31 refers to IAS 8
Accounting Policies, Changes in Accounting Estimates and
Errors, which requires that a voluntary change in accounting
policy be made only if the change results in the financial
statements providing stakeholders reliable and more relevant
information.
Management
is of the view that the fair value measurement of investment
properties that the Company adopted in fiscal year 2017 provides a)
reliable information to the financial statement users because that
measurement regularly updates the values of the Company’s
investment properties in contrast to the historical cost basis that
remains unchanged over multiple reporting periods, except for
depreciation and impairment adjustments; and b) more relevant
information to the financial statement users because this policy
presents the Company’s stakeholders a more robust analysis of
the Company’s financial performance. As referred to in IAS
40.31, it is highly unlikely that a change from the fair value
model to the historical cost model will result in a more relevant
presentation of the value of the Company’s investment
properties.
When
the Company adopted the fair value model, that decision was based
on the rebuttable presumption that an entity can reliably measure
the fair value of an investment property on a continuing basis, as
there is no evidence that the fair value of the investment property
will not be reliably measured. Exceptional cases referred to in IAS
40.53 (when an entity first acquires an investment property; when
an existing property first becomes an investment property after a
change in use; or when measuring the value an investment property
under construction) were not applicable for the Company’s
existing shopping mall assets at June 30, 2017 (the date the
Company set for adoption of the fair value model).
Furthermore, upon changing from the historical cost to fair value
measurement, the Company evaluated the reliability of its fair
value measurements and concluded that even though this is an
accounting estimate that involves a relatively high estimation
uncertainty as it is based on significant assumptions, the
measurements were reliable at initial recognition and on a
continuing basis based on the following factors:
Since there had been no recent observable comparable shopping mall
transactions in a number of years, the Company chose an
alternative, acceptable methodology to measure fair value. The
Company determined that the best estimate of the fair value of the
Company’s shopping mall properties was the present value of
the expected net cash flows that the shopping mall properties are
projected to generate. To that effect, the Company determined that
the discounted cash flow (“DCF”) model was a widely
accepted methodology by market participants in Argentina to
estimate the fair value of real estate properties as well as
other non-financial assets that generate future cash flows on a
reliable basis.
The Company considered that although the methodology relied on a
number of unobservable inputs, management believed that it could
obtain those inputs reliably from internally generated analysis and
sources, and where possible it maximized the use of relevant
observable inputs.
The principal assumptions used and the assessment of their
reliability are described below:
●
The main economic driver of real growth
in shopping mall real rents is GDP growth. The Company’s
rental agreements are mainly comprised of a maximum between a fixed
monthly rent and a variable amount calculated based on a percentage
of sales of each mall tenant. Tenants’ sales are highly
correlated to economic growth. The Company evaluated reliable
historical external data to arrive at the conclusion that a high
correlation exists between historical shopping mall revenue and GDP
growth. The more the economy grows, the more consumer spending
increases, driving tenants’ sales revenue and by extension
the variable amount of rent revenue generated at the
Company’s shopping mall properties. Based on external
reliable information obtained from a third party economic
consultant regarding Argentina’s projected economic
growth, which are consistent with IMF and OECD economic
projections, the Company
was able to project net revenues attributable to each mall property
on a reliable basis. Furthermore, the Company determined that it
possessed a robust and reliable set of historical EBITDA margin
data for each mall property and therefore, was able to capture the
specificities of each property into the projected property’s
cash flows.
●
Given the lack of a relevant long-term
peso-denominated interest rate to discount future cash flows, as
there is not a relevant market for long-term corporate or sovereign
bonds in local currency, it is widely accepted practice by market
participants and appraisers in Argentina using a DCF methodology to
convert peso-denominated cash flows into US dollars prior to using
a US dollar-denominated discount rate. To such extent, the Company
determined that it was able to obtain projected US dollar/peso
exchange rates on a reliable basis from independent third parties
(Estudio Broda, a well-known economic consultant in the Argentine
market) for periods up to 30 months. In addition, and in
order to assess the reliability of the economic assumption used,
the Company validates such projections and the results received
from the Company’s external consultant by reference the
“Reporte de Expectativas de
Mercado” (Report of Market Expectations
“REM”) prepared by the Central Bank of Argentina. This
report summarizes the main macroeconomic projections for the short
and medium term of the Argentine economy (up to 2 years) provided
by a survey of local and foreign specialized analysts. The survey
is conducted during the last three business days of each calendar
month and includes each analysts’ projections of consumer
prices, the Peso-U.S. dollar exchange rate, level of economic
activity and the result of operations of the non-financial domestic
private sector. This is a reliable source of projections related to the main
economic variables and it is commonly used by market participants.
From 30 months onwards, the Company assumed that the USD / peso
exchange rate will vary in line with projected local and US
inflation. We believe this is a reasonable assumption market
participants would generally use, in the absence of economic data
for the long term.
●
The Company was able to determine a relevant US dollar discount
rate on a reliable basis. Given the lack of a relevant long-term
peso-denominated interest rate to discount future cash flows, it is
a widely accepted practice by market participants in Argentina
using a DCF methodology to use a US-dollar discount rate; which is
built up using the US Risk Free Rate as a starting point (an
observable input), and then adjusted on the basis of other
observable inputs, such as the Argentine Country Risk Premium, the
beta, equity risk premium, debt/capital weightings, cost of debt
and tax rate.
Lastly,
although the Standard does not require entities to determine the
fair value of investment properties on the basis of a valuation
made by an external expert, the Company validated its valuation
methodology and outcome with a third-party valuation report
detailed by asset from the local affiliate of Newmark Grubb, a
well-known worldwide real estate advisory firm, recognized with
relevant professional qualification and who has vast local and
international experience in the real estate industry and in the
category of the investment property being valued.
We
would like to emphasize that we consider that the volatility of the
Argentine economy is adequately captured in each update of the
valuation model through the adoption of certain relevant figures,
such as the Country Risk Premium in Argentina. However, this
situation does not undermine our ability to use the fair value
measurement on a continuing basis for our investment
properties.
2. Your prior response to comment #1 indicates that the fluctuation
in shopping mall fair value is primarily attributable to an
increase in real rents and a decrease in the discount rate (i.e.,
reflecting changes in market participant views of shopping mall
risk).
● Explain the economic driver of growth in shopping mall real
rents.
In response to the Staff’s comment, the Company respectfully
advises the Staff that the main economic driver of real growth in
shopping mall rents is GDP growth in Argentina, which directly
affects tenants’ revenue and therefore the Company’s
variable component of rental revenue (as it is generally a
percentage of tenant sales). The Company respectfully advises the
Staff that the Company evaluated reliable historical external data
to determine the existence of a high correlation between historical
shopping mall revenue and GDP growth. Also, this assumption is
supported by internal data obtained by the Company in respect of
the performance of shopping mall sales in the past. Based on
the analysis of information from 2006 to date, historical shopping
mall revenue shows a high correlation with the evolution of nominal
GDP growth.
● Explain why office properties did not experience a similar
growth in real rents.
In
response to the Staff’s comment, the Company respectfully
advises the Staff that the market of premium offices is more
tightly related to this market’s vacancy rate, which results
from the balance between office market supply and
demand.
While
office market demand is also impacted by GDP growth (through its
impact in employment) office market supply usually responds to
demand with a certain lag (usually increases when construction
costs are low in USD and the market is showing relatively low
vacancy rates).
In
Argentina, office rental agreements are denominated in USD for an
average term of three years. That means that office cash flows are
generated in dollar-terms, thus being less affected by FX
fluctuations. As a result, fair value of office properties tend to
be more stable in USD, which is why this segment has historically
attracted investors who seek to protect value by investing in
non-financial assets.
In comparison with other markets where both office and
shopping mall rents are transacted in the same currency (and so are
affected by FX fluctuations in the same way), cash flows of office
properties and shopping malls in Argentina when converted into
dollars may have different behaviors in volatile market situations.
Shopping mall properties have cash flows that are much more
volatile since those cash flows are in pesos and net cash flow is
directly related to the current economic activity at each valuation
date.
In
summary, cash flows of office properties and shopping malls are
meaningfully different. Shopping mall business is much more
volatile, generates cash flow in Argentine pesos and its net cash
flow is directly related to the economic growth, which translates
to increased percentage of variable rent that has increased over
time. Economic growth impacts demand for office space more slowly
and with a lag over time. This is different for retail where the
effects of a stronger economy are reflected in rents almost
immediately (similarly, the effects of a deteriorating economy show
up in the values of shopping malls more rapidly than in office
buildings). In addition, office property rental agreements are
generally locked for a period of time and there is not the same
level of growth or change in office properties from one period to
the next and typically no major change in value from period to
period.
● Explain why office properties did not experience a similar
contraction in discount rates.
In the
absence of a stable capital market in Argentina, this sector has
traditionally been a local refuge for long term dollarized
investors, so in periods of relative instability, cap rates remain
relatively low.
Fair
value of office properties in Argentina generates a stable cash
flow in dollars in a highly volatile economic environment;
therefore, implicit discount rates in this segment are much lower
than in other businesses. This fact is reflected in market prices
of office properties which is the fair value measurement used by
the Company for this type of asset.
3. As of June 30, 2016 and 2015, we observe that a financial data
aggregator published ARS/USD non-deliverable forward (NDF) rates
for periods of 3 and 5 years, respectively; please
explain:
·
Why
NDF rates are, or are not, relevant to the valuation of your
shopping malls.
·
If
excluding NDF rates is consistent with IFRS 13 requirement to
maximize the use of observable inputs.
·
How
the use of NDF rates would affect your fair value
measurements
In response to the Staff’s comment, the Company respectfully
advises the Staff that the Company evaluated the NDF rates
and concluded that they were not relevant to the valuation of the
Company’s shopping malls for the following
reasons:
●
The NDF rates were
not always and/or easily accessible to market participants in
Argentina due to frequent government-imposed exchange controls in
Argentina. Until December 2015, there were exchange controls
preventing companies from operating NDF overseas.
●
The inability to
settle market positions in a foreign account since the Central Bank
of Argentina did not authorize wire transfers to settle foreign
exchange open positions entered into with a counterparty in a
foreign country through an NDF transaction.
●
The NDF rates were
based on unofficial exchange rates, typically exchange rates
transacted in a market considered illegal by governmental
authorities in Argentina during times of exchange
controls.
The Company acknowledges that IFRS 13 requires companies to
maximize the use of observable inputs. However, under certain
circumstances, those observable inputs may not be relevant to the
underlying fair value. As indicated in the Basis for Conclusions to
IFRS 13, observability is not the only criterion applied when
selecting the inputs to a valuation technique. The Company
concluded that even though there were observable NDF rates, these
were not relevant for use in the valuation of its shopping malls
portfolio due to the reasons explained above. Notwithstanding the
foregoing, the Company - in response to the SEC staff request-
estimated that the use of the NDF rates would have reduced the fair
value of its shopping mall properties by 3% and 30% as of June 30,
2016 and 2015, respectively. As further described above, the
percentage difference for 2015 is significant due to the exchange
control restrictions in place at that time – which were
described above- and render the use of the NDF rates not relevant
for use in the valuation.
4. Your response cites “macroeconomic imbalances” as a
driver of shopping mall appreciation; please explain:
● If the effect on the valuation is reflective of a
shortcoming of your valuation approach, and why or why
not.
In response to the Staff’s comment, the Company respectfully
advises the Staff that the macroeconomic assumptions used by
the Company in its valuation model were consistent with market
participant expectations at each measurement date. Since the
performance of the shopping mall business is dependent on those
variables, unexpected changes in such variables impact the fair
value measurement of shopping mall properties.
Similarly,
fair value of financial assets publicly traded in a stock exchange
(such as Argentine sovereign debt and equities) have experienced
significant changes in the recent past in response to changes in
market participants expectations related to the future performance
of the Argentine economy.
In
summary, the fact that the fair value of an asset (either shopping
mall or a financial asset) changes from period to period does not
imply that the inputs were unreliable and/or there is a shortcoming
in the valuation model used.
● If the effect on shopping mall valuations is the result of
inconsistent assumptions (e.g., inconsistencies between the
inflation rates used to forecast shopping mall rents, and the
inflation rates implicit in the peso-dollar forward
rates).
In
response to the Staff’s comment, the Company respectfully
advises the Staff that the Company included in cash flow
projections different assumptions of the economic data prevailing
at the valuation date. Economic data (projected inflation and
peso-dollar forward rates) were obtained from external sources and
are considered observable inputs.
Changes
in the expected behavior of these variables from period to period
may affect shopping mall valuations and are captured at each
measurement date through the update of the inputs used in the
valuation model.
● Why those imbalances did not affect office
valuations.
As
mentioned in the last bullets of point # 2 of this response above,
fair value of office properties is not directly responsive to
short-term changes in economic variables or market participant
expectations on the development of the Argentine economy, but to
the offer / demand of leasable office space. In addition, FX
fluctuations do not significantly affect office valuations measured
in USD since office rents are agreed in USD. A different behavior
of inflation and FX does not significantly affect office valuations
in USD. Thus, their fair value is much more stable in
USD.
5. Your response notes that the risks of each specific shopping
mall are captured in your cash flow forecasts for each mall; please
explain how your cash flow forecasts capture the specific risks of
each mall.
In
response to the SEC staff’s comment, the Company respectfully
advises the staff that the fair value of each property is assessed
separately, taking into account their specific location, market
positioning, tenant mix and quality of the property.
Regarding
the specific risks, since there is no available external data to
estimate a different beta for each property, we decided not to
assume a specific discount rate for each shopping center.
Nevertheless, it is worth mentioning that we consider there are no
significant differences between our shopping malls’ risks
since:
●
most of our
shopping centers, both in the metropolitan area of Buenos Aires and
in the rest of the country’s main cities, are placed in
premium locations, targeting the higher income
populations;
●
there are no
significant differences amongst the tenant mixes of our different
shopping centers, with most of them even sharing the same
tenants;
●
none of our
properties’ rental revenues is tied to a specific anchor
tenant with a specific agreement, whose decline might imply a
downgrade of the property’s value: and
●
all our shopping
malls share the same revenue model (our rental income is typically
the higher of the amount of the fixed rent and a percentage of the
tenants’ sales).
6. How the unadjusted applicable US Sector Spread is reflective of
Argentine market participant assumptions.
In
response to the SEC staff’s comment, the Company respectfully
advises the staff that in the WACC calculation provided by the
Company in response to the Staff’s comment #4 of the SEC
letter dated November 20, 2018, it considered US sector spread
comparable from the US shopping mall industry, which is the
reference market for valuation specialists. This WACC calculation
is built-up with US assumptions but then is further adjusted by the
use of the Argentine Country Risk Premium; which is the most
relevant input to adjust assumptions for Argentine specific
risks.
Why the US statutory tax rate is reflective of Argentine shopping
mall market participant assumptions, and how market participants
expectations regarding tax law changes would or would not affect
the use of the statutory rate.
The
Company respectfully advises the Staff that it is not using the US
statutory tax rate; rather the 35% tax rate used in the model to
determine the after-tax cost of debt is the effective tax rate
prevailing in Argentina at the valuation date.
Why US debt and equity weightings are reflective of Argentine
shopping mall market participant assumptions.
In
response to the SEC staff’s comment, the Company respectfully
advises the staff that in the WACC calculation analyzed by the
Company in response to the Staff’s comment #4 of the SEC
letter dated November 20, 2018, it considered US debt and equity
weightings only for illustration purposes. Nevertheless, in the
WACC calculation used in our valuation model, we considered our own
debt and equity weighting, reflecting the historical lower
indebtedness ratios of Argentina’s blue chip companies. Due
to the lack of other public and representative players in our
market, we consider the company to be the best proxy available for
a segment market participant in Argentina.
Why your WACC estimates are lower than those implied by use US
domiciled entity assumptions.
The
Company respectfully advises the Staff that the sole purpose for
estimating a WACC using US domiciled entities’ data as the
starting point was to assess the reasonableness of the discount
rate used by the Company in its valuation model by showing that
both rates were within the same range. It shall be noted that the
discount rate built up on the basis of US domiciled entities’
data has been adjusted to reflect local risk through the use of the
country risk premium. Had the country risk premium not been
considered, the US domiciled entities’ WACC would have been
significantly lower than our WACC.
The
main driver of the difference between our WACC estimates and the
example of US domiciled entities´ data as starting point was
the assumption for the cost of debt. We decided to use the cost of
indebtedness for Argentine blue chips companies instead of taking
the Argentine sovereign debt cost plus a premium. We think this is
more accurate because the sovereign has a track record of debt
defaults while, in general, blue chips did not default on their
debt commitments with the market or otherwise managed to
restructure them before incurring in default.
Therefore,
we consider this is the best approach for building up the WACC
calculation from the perspective of a market participant in
Argentina.
* *
*
Please
do not hesitate to contact me (212 455-3066) with any questions you
may have regarding the above responses.
Very
truly yours,
By: /s/
Jaime Mercado
Jaime
Mercado
cc:
Matias
Gaivironsky
Leonardo
Magliocco
Eduardo
Loiacono
Mariano
Tomatis
David
L. Williams